├── .gitignore ├── Book ├── ch01 │ └── .keep ├── ch02 │ └── .keep ├── ch03 │ └── .keep ├── ch04 │ └── .keep ├── ch05 │ └── .keep └── ch06 │ └── .keep ├── Codes ├── appendix-A │ ├── 01_optional-python-setup-preferences │ │ ├── README.md │ │ └── figures │ │ │ ├── activate-env.png │ │ │ ├── check-pip.png │ │ │ ├── conda-install.png │ │ │ ├── download.png │ │ │ ├── miniforge-install.png │ │ │ ├── new-env.png │ │ │ └── pytorch-installer.jpg │ ├── 02_installing-python-libraries │ │ ├── README.md │ │ ├── figures │ │ │ ├── check_1.jpg │ │ │ ├── check_2.jpg │ │ │ ├── jupyter-issues.jpg │ │ │ ├── pytorch-installer.jpg │ │ │ └── watermark.jpg │ │ ├── python_environment_check.ipynb │ │ ├── python_environment_check.py │ │ └── requirements.txt │ └── 03_main-chapter-code │ │ ├── DDP-script.py │ │ ├── code-part1.ipynb │ │ ├── code-part2.ipynb │ │ └── exercise-solutions.ipynb ├── appendix-B │ └── README.md ├── ch01 │ └── README.md ├── ch02 │ ├── 01_main-chapter-code │ │ ├── README.md │ │ ├── ch02.ipynb │ │ ├── dataloader.ipynb │ │ ├── exercise-solutions.ipynb │ │ └── the-verdict.txt │ ├── 02_bonus_bytepair-encoder │ │ ├── README.md │ │ ├── bpe_openai_gpt2.py │ │ ├── compare-bpe-tiktoken.ipynb │ │ └── gpt2_model │ │ │ └── encoder.json │ ├── 03_bonus_embedding-vs-matmul │ │ ├── README.md │ │ ├── embeddings-and-linear-layers.ipynb │ │ └── images │ │ │ ├── 1.png │ │ │ ├── 2.png │ │ │ ├── 3.png │ │ │ ├── 4.png │ │ │ └── 5.png │ ├── 09_summary │ │ └── 09_summary.ipynb │ └── README.md ├── ch03 │ ├── 01_main-chapter-code │ │ ├── README.md │ │ ├── ch03.ipynb │ │ ├── exercise-solutions.ipynb │ │ ├── figures │ │ │ ├── attention-matrix.png │ │ │ ├── attention.png │ │ │ ├── dot-product.png │ │ │ ├── dropout.png │ │ │ ├── masked.png │ │ │ ├── multi-head.png │ │ │ ├── single-head.png │ │ │ ├── weight-selfattn-1.png │ │ │ ├── weight-selfattn-2.png │ │ │ ├── weight-selfattn-3.png │ │ │ └── weight-selfattn-4.png │ │ ├── multihead-attention.ipynb │ │ └── small-text-sample.txt │ └── README.md ├── ch04 │ ├── 01_main-chapter-code │ │ ├── README.md │ │ ├── ch04.ipynb │ │ ├── exercise-solutions.ipynb │ │ ├── figures │ │ │ ├── chapter-steps.webp │ │ │ ├── ffn.webp │ │ │ ├── generate-text.webp │ │ │ ├── gpt-in-out.webp │ │ │ ├── gpt.webp │ │ │ ├── iterative-gen.webp │ │ │ ├── iterative-generate.webp │ │ │ ├── layernorm.webp │ │ │ ├── layernorm2.webp │ │ │ ├── mental-model-2.webp │ │ │ ├── mental-model-3.webp │ │ │ ├── mental-model-final.webp │ │ │ ├── mental-model.webp │ │ │ ├── overview-after-ln.webp │ │ │ ├── shortcut-example.webp │ │ │ ├── transformer-block.webp │ │ │ └── use-gpt.webp │ │ ├── gpt.py │ │ └── previous_chapters.py │ └── README.md ├── ch05 │ ├── 01_main-chapter-code │ │ ├── README.md │ │ ├── ch05.ipynb │ │ ├── gpt_download.py │ │ ├── gpt_generate.py │ │ ├── gpt_train.py │ │ ├── images │ │ │ ├── img-1.webp │ │ │ ├── img-2.webp │ │ │ └── img-3.webp │ │ ├── previous_chapters.py │ │ └── tests.py │ ├── 02_alternative_weight_loading │ │ ├── README.md │ │ ├── previous_chapters.py │ │ └── weight-loading-hf-transformers.ipynb │ ├── 03_bonus_pretraining_on_gutenberg │ │ ├── README.md │ │ ├── prepare_dataset.py │ │ ├── pretraining_simple.py │ │ └── previous_chapters.py │ ├── 04_learning_rate_schedulers │ │ └── README.md │ ├── 05_bonus_hparam_tuning │ │ ├── README.md │ │ ├── hparam_search.py │ │ ├── previous_chapters.py │ │ └── the-verdict.txt │ └── README.md ├── ch06 │ ├── 01_main-chapter-code │ │ ├── README.md │ │ ├── ch06.ipynb │ │ ├── exercise-solutions.ipynb │ │ ├── gpt-class-finetune.py │ │ ├── gpt_download.py │ │ ├── previous_chapters.py │ │ └── tests.py │ ├── 02_bonus_additional-experiments │ │ ├── README.md │ │ ├── additional-experiments.py │ │ ├── gpt_download.py │ │ └── previous_chapters.py │ └── 03_bonus_imdb-classification │ │ ├── README.md │ │ ├── download-prepare-dataset.py │ │ ├── gpt_download.py │ │ ├── previous_chapters.py │ │ ├── requirements-extra.txt │ │ ├── sklearn-baseline.ipynb │ │ ├── train-bert-hf.py │ │ ├── train-gpt.py │ │ └── train-sklearn-logreg.py └── ch07 │ ├── 01_main-chapter-code │ ├── README.md │ ├── ch07.ipynb │ ├── exercise-solutions.ipynb │ ├── exercise_experiments.py │ ├── gpt_download.py │ ├── gpt_instruction_finetuning.py │ ├── instruction-data-with-response.json │ ├── instruction-data.json │ ├── load-finetuned-model.ipynb │ ├── ollama_evaluate.py │ ├── previous_chapters.py │ └── tests.py │ ├── 02_dataset-utilities │ ├── README.md │ ├── config.json │ ├── create-passive-voice-entries.ipynb │ ├── find-near-duplicates.py │ ├── instruction-examples-modified.json │ ├── instruction-examples.json │ └── requirements-extra.txt │ ├── 03_model-evaluation │ ├── README.md │ ├── config.json │ ├── eval-example-data.json │ ├── llm-instruction-eval-ollama.ipynb │ ├── llm-instruction-eval-openai.ipynb │ ├── requirements-extra.txt │ └── scores │ │ ├── correlation-analysis.ipynb │ │ ├── gpt4-model-1-response.json │ │ ├── gpt4-model-2-response.json │ │ ├── llama3-8b-model-1-response.json │ │ └── llama3-8b-model-2-response.json │ ├── 04_preference-tuning-with-dpo │ ├── README.md │ ├── create-preference-data-ollama.ipynb │ ├── dpo-from-scratch.ipynb │ ├── instruction-data-with-preference.json │ └── previous_chapters.py │ ├── 05_dataset-generation │ ├── README.md │ ├── instruction-data-llama3-7b.json │ └── llama3-ollama.ipynb │ └── README.md ├── LICENSE.txt ├── Model_Architecture_Discussions ├── .keep ├── ChatGLM3 │ ├── README.md │ ├── configuration_chatglm_full.py │ ├── glm.py │ ├── img │ │ └── img.png │ ├── quantization.py │ ├── tokenization_chatglm.py │ ├── tokenizer.model │ ├── tokenizer_config.json │ └── 加载模型权重.ipynb ├── ChatGLM4 │ ├── chatglm4-guide.ipynb │ ├── chatglm4.ipynb │ ├── configuration_chatglm.py │ ├── modeling_chatglm.py │ └── tokenization_chatglm.py ├── MiniCPM │ ├── MiniCPM.ipynb │ ├── MiniCPM.py │ ├── MiniCPMTest.ipynb │ ├── README.md │ ├── config.json │ ├── configuration_minicpm.py │ ├── generation_config.json │ ├── gitattributes │ ├── special_tokens_map.json │ ├── tokenizer.json │ ├── tokenizer.model │ └── tokenizer_config.json ├── gptj │ ├── configuration_gptj.py │ ├── gptj.ipynb │ └── modeling_gptj.py ├── img │ └── .keep ├── llama3 │ ├── LICENSE │ ├── README.md │ ├── images │ │ ├── 42.png │ │ ├── a10.png │ │ ├── afterattention.png │ │ ├── archi.png │ │ ├── attention.png │ │ ├── embeddings.png │ │ ├── finallayer.png │ │ ├── freq_cis.png │ │ ├── god.png │ │ ├── heads.png │ │ ├── implllama3_30_0.png │ │ ├── implllama3_39_0.png │ │ ├── implllama3_41_0.png │ │ ├── implllama3_42_0.png │ │ ├── implllama3_50_0.png │ │ ├── implllama3_52_0.png │ │ ├── implllama3_54_0.png │ │ ├── karpathyminbpe.png │ │ ├── keys.png │ │ ├── keys0.png │ │ ├── last_norm.png │ │ ├── mask.png │ │ ├── model.png │ │ ├── norm.png │ │ ├── norm_after.png │ │ ├── q_per_token.png │ │ ├── qkmatmul.png │ │ ├── qkv.png │ │ ├── qsplit.png │ │ ├── rms.png │ │ ├── rope.png │ │ ├── ropesplit.png │ │ ├── softmax.png │ │ ├── stacked.png │ │ ├── swiglu.png │ │ ├── tokens.png │ │ ├── v0.png │ │ ├── value.png │ │ └── weightmatrix.png │ ├── llama3-from-scratch.ipynb │ ├── params.json │ ├── params.txt │ ├── requirements.txt │ └── tokenizer.model ├── mamba │ ├── README.md │ ├── demo.ipynb │ └── model.py ├── olmo │ ├── configuration_olmo.py │ ├── modeling_olmo.py │ └── olmo.ipynb ├── openelm │ ├── configuration_openelm.py │ ├── modeling_openelm.py │ └── openelm.ipynb ├── pangu │ ├── configuration_gptpangu.py │ ├── modeling_gptpangu.py │ ├── pangu.ipynb │ ├── tokenization_gptpangu.py │ └── tokenization_gptpangu_bak.py ├── phi-3 │ ├── configuration_phi3.py │ ├── modeling_phi3.py │ └── phi-3.ipynb ├── phi │ ├── configuration_phi.py │ ├── modeling_phi.py │ └── phi.ipynb ├── rwkv-compare │ ├── model_v1.py │ ├── model_v2.py │ ├── model_v3.py │ ├── model_v4.py │ ├── model_v5.py │ ├── model_v6.py │ └── readme.md ├── rwkv-v1 │ ├── model.py │ └── readme.md ├── rwkv-v2 │ ├── 20B_tokenizer.json │ ├── img │ │ └── 01.png │ ├── model.py │ ├── rwkv-v2-guide.ipynb │ └── rwkv-v2.ipynb ├── rwkv-v3 │ ├── 20B_tokenizer.json │ ├── model.py │ ├── model_run.py │ ├── rwkv-v3-guide.ipynb │ ├── rwkv-v3.ipynb │ └── utils.py ├── rwkv-v4 │ ├── 20B_tokenizer.json │ └── rwkv-v4-guide.ipynb ├── rwkv-v5 │ ├── RWKV-v5-guide.ipynb │ ├── RWKV_v5_demo.ipynb │ ├── img │ │ └── 01.png │ └── rwkv_vocab_v20230424.txt └── rwkv-v6 │ ├── RWKV-v6-guide.ipynb │ ├── RWKV_v6_demo.ipynb │ ├── img │ └── 01.png │ └── rwkv_vocab_v20230424.txt ├── README.md ├── Translated_Book ├── ch01 │ ├── .keep │ ├── 1.0理解大型语言模型.md │ ├── 1.1什么是LLM.md │ ├── 1.2LLMs的应用.md │ ├── 1.5利用大型数据集.ipynb │ ├── 1.6深入剖析GPT架构.ipynb │ ├── 1.7构建大语言模型.ipynb │ ├── 1.8总结.ipynb │ └── welcome.ipynb ├── ch02 │ ├── .keep │ ├── 2.1理解词嵌入.ipynb │ ├── 2.2文本分词(序列化).ipynb │ ├── 2.3将令牌转换为令牌 ID.ipynb │ ├── 2.4添加特殊上下文tokens.ipynb │ ├── 2.5 字节对编码(BPE).ipynb │ ├── 2.6使用滑动窗口进行数据采样.ipynb │ ├── 2.7 构建词符嵌入.ipynb │ ├── 2.8词位置编码.ipynb │ └── 2.文本数据处理.ipynb ├── ch03 │ ├── .keep │ ├── 3.1.ipynb │ ├── 3.2.ipynb │ ├── 3.3.ipynb │ ├── 3.4.ipynb │ ├── 3.5.ipynb │ ├── 3.6.ipynb │ └── 3.7.ipynb ├── ch04 │ ├── .keep │ ├── 4.1 从头开始实现 GPT 模型以生成文本.ipynb │ ├── 4.1.ipynb │ ├── 4.2 使用层归一化对激活进行归一化.ipynb │ ├── 4.2.ipynb │ ├── 4.3 实现使用 GELU 激活函数的前馈网络.ipynb │ ├── 4.4 增加快捷链接.ipynb │ ├── 4.5 在transfomer模块中连接注意力层和线性层.ipynb │ ├── 4.6 编码GPT模型-Copy1.ipynb │ ├── 4.6 编码GPT模型.ipynb │ └── 4.7 生成文本.ipynb ├── ch05 │ ├── .keep │ ├── 5.1 在未标记的数据上进行预训练.ipynb │ ├── 5.2.ipynb │ └── 5.3.ipynb └── img │ ├── .keep │ ├── Figure 1.1.png │ ├── Figure 1.2.png │ ├── Figure 1.3.png │ ├── Figure 1.4.png │ ├── Figure 1.5.png │ ├── Figure 1.6.png │ ├── cover-1.jpg │ ├── cover-2.jpg │ ├── fig-1-1.jpg │ ├── fig-1-2.jpg │ ├── fig-1-3.jpg │ ├── fig-1-4.jpg │ ├── fig-1-5.jpg │ ├── fig-1-6.png │ ├── fig-1-7.jpg │ ├── fig-1-8.jpg │ ├── fig-1-9.jpg │ ├── fig-1.7-1.jpg │ ├── fig-2-1.jpg │ ├── fig-2-10.jpg │ ├── fig-2-11.jpg │ ├── fig-2-12.jpg │ ├── fig-2-13.jpg │ ├── fig-2-14.jpg │ ├── fig-2-15.jpg │ ├── fig-2-16.jpg │ ├── fig-2-17.jpg │ ├── fig-2-18.jpg │ ├── fig-2-19.jpg │ ├── fig-2-2.jpg │ ├── fig-2-20.png │ ├── fig-2-21.png │ ├── fig-2-3.jpg │ ├── fig-2-4.jpg │ ├── fig-2-5.jpg │ ├── fig-2-6.jpg │ ├── fig-2-7.jpg │ ├── fig-2-8.jpg │ ├── fig-2-9.jpg │ ├── fig-3-1.jpg │ ├── fig-3-1.png │ ├── fig-3-10.jpg │ ├── fig-3-11.jpg │ ├── fig-3-12.jpg │ ├── fig-3-13.jpg │ ├── fig-3-14.jpg │ ├── fig-3-15.jpg │ ├── fig-3-16.jpg │ ├── fig-3-17.jpg │ ├── fig-3-18.jpg │ ├── fig-3-19.jpg │ ├── fig-3-2.jpg │ ├── fig-3-2.png │ ├── fig-3-20.jpg │ ├── fig-3-21.jpg │ ├── fig-3-22.jpg │ ├── fig-3-23.jpg │ ├── fig-3-24.jpg │ ├── fig-3-25.jpg │ ├── fig-3-26.jpg │ ├── fig-3-3.jpg │ ├── fig-3-3.png │ ├── fig-3-4.jpg │ ├── fig-3-4.png │ ├── fig-3-5.jpg │ ├── fig-3-5.png │ ├── fig-3-6.jpg │ ├── fig-3-6.png │ ├── fig-3-7.jpg │ ├── fig-3-8.jpg │ ├── fig-3-9.jpg │ ├── fig-4-1.jpg │ ├── fig-4-1.png │ ├── fig-4-10.jpg │ ├── fig-4-11.jpg │ ├── fig-4-12.jpg │ ├── fig-4-13.jpg │ ├── fig-4-14.jpg │ ├── fig-4-15.jpg │ ├── fig-4-16.jpg │ ├── fig-4-17.jpg │ ├── fig-4-18.jpg │ ├── fig-4-2.jpg │ ├── fig-4-2.png │ ├── fig-4-3.jpg │ ├── fig-4-3.png │ ├── fig-4-4.jpg │ ├── fig-4-4.png │ ├── fig-4-5.jpg │ ├── fig-4-5.png │ ├── fig-4-6.jpg │ ├── fig-4-6.png │ ├── fig-4-7.jpg │ ├── fig-4-7.png │ ├── fig-4-8.jpg │ ├── fig-4-9.jpg │ ├── fig-5-1.jpg │ ├── fig-5-10.jpg │ ├── fig-5-11.jpg │ ├── fig-5-11.png │ ├── fig-5-12.jpg │ ├── fig-5-12.png │ ├── fig-5-13.jpg │ ├── fig-5-13.png │ ├── fig-5-14.jpg │ ├── fig-5-15.jpg │ ├── fig-5-16.jpg │ ├── fig-5-17.jpg │ ├── fig-5-2.jpg │ ├── fig-5-3.png │ ├── fig-5-4.jpg │ ├── fig-5-5.png │ ├── fig-5-6.jpg │ ├── fig-5-7.jpg │ ├── fig-5-8.jpg │ ├── fig-5-9.jpg │ ├── fig-A-1.jpg │ ├── fig-A-10.jpg │ ├── fig-A-11.jpg │ ├── fig-A-12.jpg │ ├── fig-A-13.jpg │ ├── fig-A-2.jpg │ ├── fig-A-3.jpg │ ├── fig-A-4.jpg │ ├── fig-A-5.jpg │ ├── fig-A-6.jpg │ ├── fig-A-7.jpg │ ├── fig-A-8.jpg │ ├── fig-A-9.jpg │ ├── fig-D-1.jpg │ └── fig-D-2.jpg └── images ├── cover.jpg └── mental-model.jpg /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/.gitignore -------------------------------------------------------------------------------- /Book/ch01/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Book/ch02/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Book/ch03/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Book/ch04/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Book/ch05/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Book/ch06/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Codes/appendix-A/01_optional-python-setup-preferences/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/01_optional-python-setup-preferences/README.md -------------------------------------------------------------------------------- /Codes/appendix-A/01_optional-python-setup-preferences/figures/activate-env.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/01_optional-python-setup-preferences/figures/activate-env.png -------------------------------------------------------------------------------- /Codes/appendix-A/01_optional-python-setup-preferences/figures/check-pip.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/01_optional-python-setup-preferences/figures/check-pip.png -------------------------------------------------------------------------------- /Codes/appendix-A/01_optional-python-setup-preferences/figures/conda-install.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/01_optional-python-setup-preferences/figures/conda-install.png -------------------------------------------------------------------------------- /Codes/appendix-A/01_optional-python-setup-preferences/figures/download.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/01_optional-python-setup-preferences/figures/download.png -------------------------------------------------------------------------------- /Codes/appendix-A/01_optional-python-setup-preferences/figures/miniforge-install.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/01_optional-python-setup-preferences/figures/miniforge-install.png -------------------------------------------------------------------------------- /Codes/appendix-A/01_optional-python-setup-preferences/figures/new-env.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/01_optional-python-setup-preferences/figures/new-env.png -------------------------------------------------------------------------------- /Codes/appendix-A/01_optional-python-setup-preferences/figures/pytorch-installer.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/01_optional-python-setup-preferences/figures/pytorch-installer.jpg -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/README.md -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/figures/check_1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/figures/check_1.jpg -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/figures/check_2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/figures/check_2.jpg -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/figures/jupyter-issues.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/figures/jupyter-issues.jpg -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/figures/pytorch-installer.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/figures/pytorch-installer.jpg -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/figures/watermark.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/figures/watermark.jpg -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/python_environment_check.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/python_environment_check.ipynb -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/python_environment_check.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/python_environment_check.py -------------------------------------------------------------------------------- /Codes/appendix-A/02_installing-python-libraries/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/02_installing-python-libraries/requirements.txt -------------------------------------------------------------------------------- /Codes/appendix-A/03_main-chapter-code/DDP-script.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/03_main-chapter-code/DDP-script.py -------------------------------------------------------------------------------- /Codes/appendix-A/03_main-chapter-code/code-part1.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/03_main-chapter-code/code-part1.ipynb -------------------------------------------------------------------------------- /Codes/appendix-A/03_main-chapter-code/code-part2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/03_main-chapter-code/code-part2.ipynb -------------------------------------------------------------------------------- /Codes/appendix-A/03_main-chapter-code/exercise-solutions.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-A/03_main-chapter-code/exercise-solutions.ipynb -------------------------------------------------------------------------------- /Codes/appendix-B/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/appendix-B/README.md -------------------------------------------------------------------------------- /Codes/ch01/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch01/README.md -------------------------------------------------------------------------------- /Codes/ch02/01_main-chapter-code/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/01_main-chapter-code/README.md -------------------------------------------------------------------------------- /Codes/ch02/01_main-chapter-code/ch02.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/01_main-chapter-code/ch02.ipynb -------------------------------------------------------------------------------- /Codes/ch02/01_main-chapter-code/dataloader.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/01_main-chapter-code/dataloader.ipynb -------------------------------------------------------------------------------- /Codes/ch02/01_main-chapter-code/exercise-solutions.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/01_main-chapter-code/exercise-solutions.ipynb -------------------------------------------------------------------------------- /Codes/ch02/01_main-chapter-code/the-verdict.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/01_main-chapter-code/the-verdict.txt -------------------------------------------------------------------------------- /Codes/ch02/02_bonus_bytepair-encoder/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/02_bonus_bytepair-encoder/README.md -------------------------------------------------------------------------------- /Codes/ch02/02_bonus_bytepair-encoder/bpe_openai_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/02_bonus_bytepair-encoder/bpe_openai_gpt2.py -------------------------------------------------------------------------------- /Codes/ch02/02_bonus_bytepair-encoder/compare-bpe-tiktoken.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/02_bonus_bytepair-encoder/compare-bpe-tiktoken.ipynb -------------------------------------------------------------------------------- /Codes/ch02/02_bonus_bytepair-encoder/gpt2_model/encoder.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/02_bonus_bytepair-encoder/gpt2_model/encoder.json -------------------------------------------------------------------------------- /Codes/ch02/03_bonus_embedding-vs-matmul/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/03_bonus_embedding-vs-matmul/README.md -------------------------------------------------------------------------------- /Codes/ch02/03_bonus_embedding-vs-matmul/embeddings-and-linear-layers.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/03_bonus_embedding-vs-matmul/embeddings-and-linear-layers.ipynb -------------------------------------------------------------------------------- /Codes/ch02/03_bonus_embedding-vs-matmul/images/1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/03_bonus_embedding-vs-matmul/images/1.png -------------------------------------------------------------------------------- /Codes/ch02/03_bonus_embedding-vs-matmul/images/2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/03_bonus_embedding-vs-matmul/images/2.png -------------------------------------------------------------------------------- /Codes/ch02/03_bonus_embedding-vs-matmul/images/3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/03_bonus_embedding-vs-matmul/images/3.png -------------------------------------------------------------------------------- /Codes/ch02/03_bonus_embedding-vs-matmul/images/4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/03_bonus_embedding-vs-matmul/images/4.png -------------------------------------------------------------------------------- /Codes/ch02/03_bonus_embedding-vs-matmul/images/5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/03_bonus_embedding-vs-matmul/images/5.png -------------------------------------------------------------------------------- /Codes/ch02/09_summary/09_summary.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/09_summary/09_summary.ipynb -------------------------------------------------------------------------------- /Codes/ch02/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch02/README.md -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/README.md -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/ch03.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/ch03.ipynb -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/exercise-solutions.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/exercise-solutions.ipynb -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/attention-matrix.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/attention-matrix.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/attention.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/attention.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/dot-product.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/dot-product.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/dropout.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/dropout.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/masked.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/masked.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/multi-head.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/multi-head.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/single-head.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/single-head.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/weight-selfattn-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/weight-selfattn-1.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/weight-selfattn-2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/weight-selfattn-2.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/weight-selfattn-3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/weight-selfattn-3.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/figures/weight-selfattn-4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/figures/weight-selfattn-4.png -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/multihead-attention.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/multihead-attention.ipynb -------------------------------------------------------------------------------- /Codes/ch03/01_main-chapter-code/small-text-sample.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/01_main-chapter-code/small-text-sample.txt -------------------------------------------------------------------------------- /Codes/ch03/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch03/README.md -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/README.md -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/ch04.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/ch04.ipynb -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/exercise-solutions.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/exercise-solutions.ipynb -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/chapter-steps.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/chapter-steps.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/ffn.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/ffn.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/generate-text.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/generate-text.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/gpt-in-out.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/gpt-in-out.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/gpt.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/gpt.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/iterative-gen.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/iterative-gen.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/iterative-generate.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/iterative-generate.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/layernorm.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/layernorm.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/layernorm2.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/layernorm2.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/mental-model-2.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/mental-model-2.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/mental-model-3.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/mental-model-3.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/mental-model-final.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/mental-model-final.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/mental-model.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/mental-model.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/overview-after-ln.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/overview-after-ln.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/shortcut-example.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/shortcut-example.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/transformer-block.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/transformer-block.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/figures/use-gpt.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/figures/use-gpt.webp -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/gpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/gpt.py -------------------------------------------------------------------------------- /Codes/ch04/01_main-chapter-code/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/01_main-chapter-code/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch04/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch04/README.md -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/README.md -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/ch05.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/ch05.ipynb -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/gpt_download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/gpt_download.py -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/gpt_generate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/gpt_generate.py -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/gpt_train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/gpt_train.py -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/images/img-1.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/images/img-1.webp -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/images/img-2.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/images/img-2.webp -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/images/img-3.webp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/images/img-3.webp -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch05/01_main-chapter-code/tests.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/01_main-chapter-code/tests.py -------------------------------------------------------------------------------- /Codes/ch05/02_alternative_weight_loading/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/02_alternative_weight_loading/README.md -------------------------------------------------------------------------------- /Codes/ch05/02_alternative_weight_loading/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/02_alternative_weight_loading/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch05/02_alternative_weight_loading/weight-loading-hf-transformers.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/02_alternative_weight_loading/weight-loading-hf-transformers.ipynb -------------------------------------------------------------------------------- /Codes/ch05/03_bonus_pretraining_on_gutenberg/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/03_bonus_pretraining_on_gutenberg/README.md -------------------------------------------------------------------------------- /Codes/ch05/03_bonus_pretraining_on_gutenberg/prepare_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/03_bonus_pretraining_on_gutenberg/prepare_dataset.py -------------------------------------------------------------------------------- /Codes/ch05/03_bonus_pretraining_on_gutenberg/pretraining_simple.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/03_bonus_pretraining_on_gutenberg/pretraining_simple.py -------------------------------------------------------------------------------- /Codes/ch05/03_bonus_pretraining_on_gutenberg/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/03_bonus_pretraining_on_gutenberg/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch05/04_learning_rate_schedulers/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/04_learning_rate_schedulers/README.md -------------------------------------------------------------------------------- /Codes/ch05/05_bonus_hparam_tuning/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/05_bonus_hparam_tuning/README.md -------------------------------------------------------------------------------- /Codes/ch05/05_bonus_hparam_tuning/hparam_search.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/05_bonus_hparam_tuning/hparam_search.py -------------------------------------------------------------------------------- /Codes/ch05/05_bonus_hparam_tuning/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/05_bonus_hparam_tuning/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch05/05_bonus_hparam_tuning/the-verdict.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/05_bonus_hparam_tuning/the-verdict.txt -------------------------------------------------------------------------------- /Codes/ch05/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch05/README.md -------------------------------------------------------------------------------- /Codes/ch06/01_main-chapter-code/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/01_main-chapter-code/README.md -------------------------------------------------------------------------------- /Codes/ch06/01_main-chapter-code/ch06.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/01_main-chapter-code/ch06.ipynb -------------------------------------------------------------------------------- /Codes/ch06/01_main-chapter-code/exercise-solutions.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/01_main-chapter-code/exercise-solutions.ipynb -------------------------------------------------------------------------------- /Codes/ch06/01_main-chapter-code/gpt-class-finetune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/01_main-chapter-code/gpt-class-finetune.py -------------------------------------------------------------------------------- /Codes/ch06/01_main-chapter-code/gpt_download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/01_main-chapter-code/gpt_download.py -------------------------------------------------------------------------------- /Codes/ch06/01_main-chapter-code/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/01_main-chapter-code/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch06/01_main-chapter-code/tests.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/01_main-chapter-code/tests.py -------------------------------------------------------------------------------- /Codes/ch06/02_bonus_additional-experiments/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/02_bonus_additional-experiments/README.md -------------------------------------------------------------------------------- /Codes/ch06/02_bonus_additional-experiments/additional-experiments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/02_bonus_additional-experiments/additional-experiments.py -------------------------------------------------------------------------------- /Codes/ch06/02_bonus_additional-experiments/gpt_download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/02_bonus_additional-experiments/gpt_download.py -------------------------------------------------------------------------------- /Codes/ch06/02_bonus_additional-experiments/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/02_bonus_additional-experiments/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/03_bonus_imdb-classification/README.md -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/download-prepare-dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/03_bonus_imdb-classification/download-prepare-dataset.py -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/gpt_download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/03_bonus_imdb-classification/gpt_download.py -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/03_bonus_imdb-classification/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/requirements-extra.txt: -------------------------------------------------------------------------------- 1 | transformers>=4.33.2 2 | scikit-learn>=1.3.0 -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/sklearn-baseline.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/03_bonus_imdb-classification/sklearn-baseline.ipynb -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/train-bert-hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/03_bonus_imdb-classification/train-bert-hf.py -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/train-gpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/03_bonus_imdb-classification/train-gpt.py -------------------------------------------------------------------------------- /Codes/ch06/03_bonus_imdb-classification/train-sklearn-logreg.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch06/03_bonus_imdb-classification/train-sklearn-logreg.py -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/README.md -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/ch07.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/ch07.ipynb -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/exercise-solutions.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/exercise-solutions.ipynb -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/exercise_experiments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/exercise_experiments.py -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/gpt_download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/gpt_download.py -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/gpt_instruction_finetuning.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/gpt_instruction_finetuning.py -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/instruction-data-with-response.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/instruction-data-with-response.json -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/instruction-data.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/instruction-data.json -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/load-finetuned-model.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/load-finetuned-model.ipynb -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/ollama_evaluate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/ollama_evaluate.py -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch07/01_main-chapter-code/tests.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/01_main-chapter-code/tests.py -------------------------------------------------------------------------------- /Codes/ch07/02_dataset-utilities/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/02_dataset-utilities/README.md -------------------------------------------------------------------------------- /Codes/ch07/02_dataset-utilities/config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/02_dataset-utilities/config.json -------------------------------------------------------------------------------- /Codes/ch07/02_dataset-utilities/create-passive-voice-entries.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/02_dataset-utilities/create-passive-voice-entries.ipynb -------------------------------------------------------------------------------- /Codes/ch07/02_dataset-utilities/find-near-duplicates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/02_dataset-utilities/find-near-duplicates.py -------------------------------------------------------------------------------- /Codes/ch07/02_dataset-utilities/instruction-examples-modified.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/02_dataset-utilities/instruction-examples-modified.json -------------------------------------------------------------------------------- /Codes/ch07/02_dataset-utilities/instruction-examples.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/02_dataset-utilities/instruction-examples.json -------------------------------------------------------------------------------- /Codes/ch07/02_dataset-utilities/requirements-extra.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/02_dataset-utilities/requirements-extra.txt -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/README.md -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/config.json -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/eval-example-data.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/eval-example-data.json -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/llm-instruction-eval-ollama.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/llm-instruction-eval-ollama.ipynb -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/llm-instruction-eval-openai.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/llm-instruction-eval-openai.ipynb -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/requirements-extra.txt: -------------------------------------------------------------------------------- 1 | openai>=1.30.3 2 | tqdm>=4.65.0 3 | -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/scores/correlation-analysis.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/scores/correlation-analysis.ipynb -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/scores/gpt4-model-1-response.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/scores/gpt4-model-1-response.json -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/scores/gpt4-model-2-response.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/scores/gpt4-model-2-response.json -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/scores/llama3-8b-model-1-response.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/scores/llama3-8b-model-1-response.json -------------------------------------------------------------------------------- /Codes/ch07/03_model-evaluation/scores/llama3-8b-model-2-response.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/03_model-evaluation/scores/llama3-8b-model-2-response.json -------------------------------------------------------------------------------- /Codes/ch07/04_preference-tuning-with-dpo/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/04_preference-tuning-with-dpo/README.md -------------------------------------------------------------------------------- /Codes/ch07/04_preference-tuning-with-dpo/create-preference-data-ollama.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/04_preference-tuning-with-dpo/create-preference-data-ollama.ipynb -------------------------------------------------------------------------------- /Codes/ch07/04_preference-tuning-with-dpo/dpo-from-scratch.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/04_preference-tuning-with-dpo/dpo-from-scratch.ipynb -------------------------------------------------------------------------------- /Codes/ch07/04_preference-tuning-with-dpo/instruction-data-with-preference.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/04_preference-tuning-with-dpo/instruction-data-with-preference.json -------------------------------------------------------------------------------- /Codes/ch07/04_preference-tuning-with-dpo/previous_chapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/04_preference-tuning-with-dpo/previous_chapters.py -------------------------------------------------------------------------------- /Codes/ch07/05_dataset-generation/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/05_dataset-generation/README.md -------------------------------------------------------------------------------- /Codes/ch07/05_dataset-generation/instruction-data-llama3-7b.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/05_dataset-generation/instruction-data-llama3-7b.json -------------------------------------------------------------------------------- /Codes/ch07/05_dataset-generation/llama3-ollama.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/05_dataset-generation/llama3-ollama.ipynb -------------------------------------------------------------------------------- /Codes/ch07/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Codes/ch07/README.md -------------------------------------------------------------------------------- /LICENSE.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/LICENSE.txt -------------------------------------------------------------------------------- /Model_Architecture_Discussions/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/README.md -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/configuration_chatglm_full.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/configuration_chatglm_full.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/glm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/glm.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/img/img.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/img/img.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/quantization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/quantization.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/tokenization_chatglm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/tokenization_chatglm.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/tokenizer.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/tokenizer.model -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/tokenizer_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/tokenizer_config.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM3/加载模型权重.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM3/加载模型权重.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM4/chatglm4-guide.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM4/chatglm4-guide.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM4/chatglm4.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM4/chatglm4.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM4/configuration_chatglm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM4/configuration_chatglm.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM4/modeling_chatglm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM4/modeling_chatglm.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/ChatGLM4/tokenization_chatglm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/ChatGLM4/tokenization_chatglm.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/MiniCPM.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/MiniCPM.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/MiniCPM.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/MiniCPM.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/MiniCPMTest.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/MiniCPMTest.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/README.md -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/config.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/configuration_minicpm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/configuration_minicpm.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/generation_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/generation_config.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/gitattributes: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/gitattributes -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/special_tokens_map.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/special_tokens_map.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/tokenizer.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/tokenizer.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/tokenizer.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/tokenizer.model -------------------------------------------------------------------------------- /Model_Architecture_Discussions/MiniCPM/tokenizer_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/MiniCPM/tokenizer_config.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/gptj/configuration_gptj.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/gptj/configuration_gptj.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/gptj/gptj.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/gptj/gptj.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/gptj/modeling_gptj.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/gptj/modeling_gptj.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/img/.keep: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/LICENSE -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/README.md -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/42.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/42.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/a10.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/a10.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/afterattention.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/afterattention.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/archi.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/archi.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/attention.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/attention.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/embeddings.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/embeddings.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/finallayer.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/finallayer.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/freq_cis.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/freq_cis.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/god.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/god.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/heads.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/heads.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/implllama3_30_0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/implllama3_30_0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/implllama3_39_0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/implllama3_39_0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/implllama3_41_0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/implllama3_41_0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/implllama3_42_0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/implllama3_42_0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/implllama3_50_0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/implllama3_50_0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/implllama3_52_0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/implllama3_52_0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/implllama3_54_0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/implllama3_54_0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/karpathyminbpe.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/karpathyminbpe.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/keys.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/keys.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/keys0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/keys0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/last_norm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/last_norm.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/mask.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/mask.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/model.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/model.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/norm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/norm.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/norm_after.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/norm_after.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/q_per_token.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/q_per_token.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/qkmatmul.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/qkmatmul.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/qkv.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/qkv.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/qsplit.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/qsplit.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/rms.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/rms.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/rope.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/rope.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/ropesplit.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/ropesplit.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/softmax.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/softmax.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/stacked.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/stacked.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/swiglu.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/swiglu.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/tokens.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/tokens.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/v0.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/v0.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/value.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/value.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/images/weightmatrix.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/images/weightmatrix.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/llama3-from-scratch.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/llama3-from-scratch.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/params.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/params.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/params.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/params.txt -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/requirements.txt -------------------------------------------------------------------------------- /Model_Architecture_Discussions/llama3/tokenizer.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/llama3/tokenizer.model -------------------------------------------------------------------------------- /Model_Architecture_Discussions/mamba/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/mamba/README.md -------------------------------------------------------------------------------- /Model_Architecture_Discussions/mamba/demo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/mamba/demo.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/mamba/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/mamba/model.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/olmo/configuration_olmo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/olmo/configuration_olmo.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/olmo/modeling_olmo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/olmo/modeling_olmo.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/olmo/olmo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/olmo/olmo.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/openelm/configuration_openelm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/openelm/configuration_openelm.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/openelm/modeling_openelm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/openelm/modeling_openelm.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/openelm/openelm.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/openelm/openelm.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/pangu/configuration_gptpangu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/pangu/configuration_gptpangu.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/pangu/modeling_gptpangu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/pangu/modeling_gptpangu.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/pangu/pangu.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/pangu/pangu.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/pangu/tokenization_gptpangu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/pangu/tokenization_gptpangu.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/pangu/tokenization_gptpangu_bak.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/pangu/tokenization_gptpangu_bak.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/phi-3/configuration_phi3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/phi-3/configuration_phi3.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/phi-3/modeling_phi3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/phi-3/modeling_phi3.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/phi-3/phi-3.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/phi-3/phi-3.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/phi/configuration_phi.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/phi/configuration_phi.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/phi/modeling_phi.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/phi/modeling_phi.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/phi/phi.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/phi/phi.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-compare/model_v1.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-compare/model_v1.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-compare/model_v2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-compare/model_v2.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-compare/model_v3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-compare/model_v3.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-compare/model_v4.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-compare/model_v4.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-compare/model_v5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-compare/model_v5.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-compare/model_v6.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-compare/model_v6.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-compare/readme.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-compare/readme.md -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v1/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v1/model.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v1/readme.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v1/readme.md -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v2/20B_tokenizer.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v2/20B_tokenizer.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v2/img/01.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v2/img/01.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v2/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v2/model.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v2/rwkv-v2-guide.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v2/rwkv-v2-guide.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v2/rwkv-v2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v2/rwkv-v2.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v3/20B_tokenizer.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v3/20B_tokenizer.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v3/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v3/model.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v3/model_run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v3/model_run.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v3/rwkv-v3-guide.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v3/rwkv-v3-guide.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v3/rwkv-v3.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v3/rwkv-v3.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v3/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v3/utils.py -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v4/20B_tokenizer.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v4/20B_tokenizer.json -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v4/rwkv-v4-guide.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v4/rwkv-v4-guide.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v5/RWKV-v5-guide.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v5/RWKV-v5-guide.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v5/RWKV_v5_demo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v5/RWKV_v5_demo.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v5/img/01.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v5/img/01.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v5/rwkv_vocab_v20230424.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v5/rwkv_vocab_v20230424.txt -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v6/RWKV-v6-guide.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v6/RWKV-v6-guide.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v6/RWKV_v6_demo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v6/RWKV_v6_demo.ipynb -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v6/img/01.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v6/img/01.png -------------------------------------------------------------------------------- /Model_Architecture_Discussions/rwkv-v6/rwkv_vocab_v20230424.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Model_Architecture_Discussions/rwkv-v6/rwkv_vocab_v20230424.txt -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/README.md -------------------------------------------------------------------------------- /Translated_Book/ch01/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Translated_Book/ch01/1.0理解大型语言模型.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch01/1.0理解大型语言模型.md -------------------------------------------------------------------------------- /Translated_Book/ch01/1.1什么是LLM.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch01/1.1什么是LLM.md -------------------------------------------------------------------------------- /Translated_Book/ch01/1.2LLMs的应用.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch01/1.2LLMs的应用.md -------------------------------------------------------------------------------- /Translated_Book/ch01/1.5利用大型数据集.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch01/1.5利用大型数据集.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch01/1.6深入剖析GPT架构.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch01/1.6深入剖析GPT架构.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch01/1.7构建大语言模型.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch01/1.7构建大语言模型.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch01/1.8总结.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch01/1.8总结.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch01/welcome.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch01/welcome.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Translated_Book/ch02/2.1理解词嵌入.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.1理解词嵌入.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/2.2文本分词(序列化).ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.2文本分词(序列化).ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/2.3将令牌转换为令牌 ID.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.3将令牌转换为令牌 ID.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/2.4添加特殊上下文tokens.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.4添加特殊上下文tokens.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/2.5 字节对编码(BPE).ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.5 字节对编码(BPE).ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/2.6使用滑动窗口进行数据采样.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.6使用滑动窗口进行数据采样.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/2.7 构建词符嵌入.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.7 构建词符嵌入.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/2.8词位置编码.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.8词位置编码.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch02/2.文本数据处理.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch02/2.文本数据处理.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch03/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Translated_Book/ch03/3.1.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch03/3.1.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch03/3.2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch03/3.2.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch03/3.3.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch03/3.3.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch03/3.4.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch03/3.4.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch03/3.5.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch03/3.5.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch03/3.6.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch03/3.6.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch03/3.7.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch03/3.7.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Translated_Book/ch04/4.1 从头开始实现 GPT 模型以生成文本.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.1 从头开始实现 GPT 模型以生成文本.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.1.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.1.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.2 使用层归一化对激活进行归一化.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.2 使用层归一化对激活进行归一化.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.2.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.3 实现使用 GELU 激活函数的前馈网络.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.3 实现使用 GELU 激活函数的前馈网络.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.4 增加快捷链接.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.4 增加快捷链接.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.5 在transfomer模块中连接注意力层和线性层.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.5 在transfomer模块中连接注意力层和线性层.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.6 编码GPT模型-Copy1.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.6 编码GPT模型-Copy1.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.6 编码GPT模型.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.6 编码GPT模型.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch04/4.7 生成文本.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch04/4.7 生成文本.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch05/.keep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /Translated_Book/ch05/5.1 在未标记的数据上进行预训练.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch05/5.1 在未标记的数据上进行预训练.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch05/5.2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch05/5.2.ipynb -------------------------------------------------------------------------------- /Translated_Book/ch05/5.3.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/ch05/5.3.ipynb -------------------------------------------------------------------------------- /Translated_Book/img/.keep: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /Translated_Book/img/Figure 1.1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/Figure 1.1.png -------------------------------------------------------------------------------- /Translated_Book/img/Figure 1.2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/Figure 1.2.png -------------------------------------------------------------------------------- /Translated_Book/img/Figure 1.3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/Figure 1.3.png -------------------------------------------------------------------------------- /Translated_Book/img/Figure 1.4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/Figure 1.4.png -------------------------------------------------------------------------------- /Translated_Book/img/Figure 1.5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/Figure 1.5.png -------------------------------------------------------------------------------- /Translated_Book/img/Figure 1.6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/Figure 1.6.png -------------------------------------------------------------------------------- /Translated_Book/img/cover-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/cover-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/cover-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/cover-2.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-2.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-3.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-4.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-5.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-6.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-7.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-8.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1-9.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1-9.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-1.7-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-1.7-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-10.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-10.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-11.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-11.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-12.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-12.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-13.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-13.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-14.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-14.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-15.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-15.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-16.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-16.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-17.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-17.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-18.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-18.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-19.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-19.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-2.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-20.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-20.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-21.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-21.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-3.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-4.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-5.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-6.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-7.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-8.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-2-9.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-2-9.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-1.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-10.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-10.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-11.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-11.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-12.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-12.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-13.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-13.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-14.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-14.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-15.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-15.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-16.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-16.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-17.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-17.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-18.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-18.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-19.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-19.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-2.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-2.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-20.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-20.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-21.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-21.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-22.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-22.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-23.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-23.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-24.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-24.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-25.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-25.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-26.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-26.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-3.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-3.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-4.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-4.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-5.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-5.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-6.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-6.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-7.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-8.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-3-9.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-3-9.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-1.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-10.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-10.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-11.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-11.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-12.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-12.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-13.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-13.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-14.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-14.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-15.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-15.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-16.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-16.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-17.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-17.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-18.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-18.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-2.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-2.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-3.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-3.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-4.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-4.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-5.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-5.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-6.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-6.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-7.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-7.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-7.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-8.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-4-9.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-4-9.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-10.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-10.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-11.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-11.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-11.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-11.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-12.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-12.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-12.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-12.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-13.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-13.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-13.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-13.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-14.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-14.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-15.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-15.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-16.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-16.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-17.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-17.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-2.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-3.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-4.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-5.png -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-6.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-7.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-8.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-5-9.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-5-9.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-10.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-10.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-11.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-11.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-12.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-12.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-13.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-13.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-2.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-3.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-3.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-4.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-4.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-5.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-5.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-6.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-6.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-7.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-7.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-8.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-8.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-A-9.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-A-9.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-D-1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-D-1.jpg -------------------------------------------------------------------------------- /Translated_Book/img/fig-D-2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/Translated_Book/img/fig-D-2.jpg -------------------------------------------------------------------------------- /images/cover.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/images/cover.jpg -------------------------------------------------------------------------------- /images/mental-model.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/datawhalechina/llms-from-scratch-cn/HEAD/images/mental-model.jpg --------------------------------------------------------------------------------