├── .gitignore ├── README.md ├── Stanford_cs229.ipynb ├── illya_NeurIPS_2024.ipynb ├── imgs ├── 2880px-MCTS_(English).svg.png ├── 4d_rotation.png ├── accelerate.png ├── adap_kl_coef.png ├── agent_rl.png ├── alpaca-instruct.png ├── bf-float.png ├── bloom.png ├── bloom_lora.png ├── bpe.png ├── c4.png ├── cg.png ├── ch-dataset.png ├── cope.png ├── decoding-qkv.png ├── drl.png ├── en-dataset.png ├── float16.png ├── floats-rep.png ├── generating.png ├── gpt-2-autoregression-2.gif ├── gpt2_bert_training.png ├── gpt4-moe.png ├── gpt4-price.png ├── gpt_config.jpeg ├── gqa.png ├── grad-accumulation.png ├── hf_llama_rope.png ├── kv-cache.gif ├── kvcache-arch.png ├── kvmatrices.png ├── llama2-weights.png ├── llama_weights.png ├── llm-archs.png ├── llm-int8.png ├── llm-tokenizer.png ├── lora-weights.png ├── lora.png ├── meta-llama-model.png ├── model-base-model-free.png ├── model_clm.png ├── model_mlm.png ├── model_seq2seq.png ├── moe.png ├── openrlhf.png ├── peft.png ├── pile.png ├── ppo_algo.png ├── qkv.png ├── quant-freeze.png ├── regular-finetuning-alt.png ├── regular-finetuning.png ├── rel_position_matrix.png ├── rlhf.png ├── rm_ppo.png ├── rope_1.png ├── rope_2.png ├── rope_paper.png ├── self-instruct.png ├── t+1.png ├── token-kvcache.png ├── token_steps.png ├── trl-sft.png ├── trl_overview.png ├── trl_pipeline.png ├── trl_ppo_loss.png ├── trl_steps.png ├── unique-tokens.jpeg └── value_iteration.png └── tutorials ├── 01_llama_intro.ipynb ├── 02_lora_baics.ipynb ├── 03_lora_practice.ipynb ├── 04_PEFT_lora.ipynb ├── 05_float16_bf16.ipynb ├── 06_alpaca_llama_7b_lora.ipynb ├── 07_fp16_mixed_precision_training.ipynb ├── Chinese(中文) └── chinese_llama2.ipynb ├── LLM.int8.ipynb ├── LLMs & NLPs.ipynb ├── alpaca_vicuna.ipynb ├── attention └── flash_attention.ipynb ├── bitsandbytes.ipynb ├── bloom_源码分析.ipynb ├── chatgpt_reviews.ipynb ├── code_sql └── overall.ipynb ├── cuda_basics.ipynb ├── discussion ├── imgs │ └── search_methods.png └── o1.ipynb ├── drl ├── 01_q_learning_dqn.ipynb ├── imgs │ ├── REINFOCE.png │ ├── ac_actor_pg.png │ ├── ac_algo.png │ ├── ac_critic_td.png │ ├── agent_as_nn.png │ ├── discounted_reward.png │ ├── game_mcts_1.png │ ├── game_mcts_2.png │ ├── game_mcts_3.png │ ├── game_mcts_4.png │ ├── her_idea.png │ ├── importance_sampling_issue.png │ ├── lm_rl.png │ ├── mab-definition.png │ ├── mab_algo.png │ ├── mc_estimate_v_pi.png │ ├── mcts_algo.png │ ├── mcts_exam_1.png │ ├── mcts_exam_2.png │ ├── mcts_exam_3.png │ ├── mcts_exam_4.png │ ├── node-edge.png │ ├── pg_gradient.png │ ├── pg_is_advantage.png │ ├── pg_loss.png │ ├── pieter_abbeel_trpo.png │ ├── policy_net_update.png │ ├── policy_net_value_net_sharing.png │ ├── policy_value_update_summary.png │ ├── ppo2.png │ ├── ppo_algo_lhy.png │ ├── ppo_clip.png │ ├── ppo_clip_curve.png │ ├── q_values.png │ ├── reinforce_monte_carlo.png │ ├── reinforce_with_baseline_approx.png │ ├── rl_setting.png │ ├── rollout.png │ ├── s_a_r.png │ ├── s_a_r_p.png │ ├── s_a_r_pi.png │ ├── suitable_credit.png │ ├── td_learning.png │ ├── td_trip.png │ ├── td_update.png │ ├── td_v_pi.png │ ├── trpo_Lagrange_duality.png │ ├── trpo_obj.png │ ├── trpo_summary.png │ ├── value_iter.png │ ├── value_loss.png │ └── value_net_update.png ├── mcts │ ├── alpha_zero.ipynb │ ├── alphago_mcts.ipynb │ ├── mcts_01_intro_bascis.ipynb │ ├── mcts_02_from_scartch.ipynb │ ├── mcts_03_binary_tree.ipynb │ ├── mcts_04_rl.ipynb │ ├── mcts_05_tic-tac-toe.ipynb │ ├── mcts_06_vis.ipynb │ ├── mcts_demo.py │ └── mcts_supp.ipynb ├── mdp_basics.ipynb ├── mdps_policy_gradient.ipynb ├── mdps_value_iteration.ipynb ├── multi_armed_bandits.ipynb ├── o1_self_rl.ipynb ├── policy_based │ ├── PPO_lihongyi_basics.ipynb │ ├── actor_critic.ipynb │ ├── annotated_trpo.ipynb │ ├── pg_trpo_ppo.ipynb │ ├── ppo_gae.ipynb │ ├── ppo_supp.ipynb │ └── trpo_optimization.ipynb ├── policy_based_rl.ipynb ├── policy_gradient_with_basline.ipynb ├── pomdp.ipynb ├── rl_basics.ipynb ├── rl_intro_review.ipynb ├── rlhf │ └── RM_loss.ipynb ├── rl应用 │ └── 供应链优化.ipynb ├── temporal_difference_bootstrapping.ipynb ├── utils │ ├── gridworld.py │ ├── mdp.py │ ├── rendering_utils.py │ └── test.ipynb ├── value_based │ ├── q_learning_lihongyi.ipynb │ └── td.ipynb ├── 对(深度)强化学习的认识与理解.ipynb └── 经典 envs.ipynb ├── evaluation ├── datasets与大模型评估(evaluation).ipynb └── test_questions.ipynb ├── falcon.ipynb ├── gpt2.ipynb ├── gpt2_train_tokenizer.ipynb ├── gpt_family.ipynb ├── gradient_accumulation.ipynb ├── gradient_checkpointing.ipynb ├── instruction_chat └── chat_template.ipynb ├── karpathy ├── 01_build_GPT.ipynb └── input.txt ├── llama ├── RewardModel.ipynb ├── attention_map_vis.ipynb ├── llama2_7b_llama3_8b.ipynb ├── llama2_tokenizer.model ├── llama3_0419.ipynb ├── llama3_supp_data_config.ipynb ├── llama3_tokenizer.model └── tokenizer.py ├── llama2_introduction.ipynb ├── llama2_src_01.ipynb ├── llama2_src_02_RMSNorm_SwiGLU.ipynb ├── llama2_src_03_RoPE.ipynb ├── llama2_src_03_RoPE_01_complex_position_encoding.ipynb ├── llama2_src_03_RoPE_apply_rotary_emb.ipynb ├── llama2_src_cache_kv.ipynb ├── llama2_src_generate.ipynb ├── llama2_src_grouped_query_attention.ipynb ├── llama2_src_text_chat.ipynb ├── llama_tokenizer.ipynb ├── llm_training_sft.ipynb ├── llms_models_datasets_finetune.ipynb ├── lora_all_parameters.ipynb ├── misc └── gradio.ipynb ├── model_finetune_quantization.ipynb ├── moe ├── Mistral_basics.ipynb ├── basics.ipynb └── mistral_src_code.ipynb ├── multi_model_vit_clip.ipynb ├── peft ├── 04_PEFT_lora.ipynb ├── PEFT_supp.ipynb ├── lora_inject.ipynb ├── lora_merge.ipynb ├── lora_svd_basics.ipynb ├── lora_vs_whole_model_1.ipynb └── lora_vs_whole_model_2.ipynb ├── position_encoding.ipynb ├── position_encoding ├── long_context_window.ipynb ├── rope-beta-encoding.ipynb └── rope_cope.ipynb ├── pretrain_corpus_大模型预训练语料.ipynb ├── quantization_gptq.ipynb ├── reward_model ├── Mistral-7b-dpo.ipynb ├── bradley_terry_modeling.ipynb ├── dpo.ipynb ├── dpo_llama2.ipynb ├── imgs │ └── quality-labels.png ├── ppo_basics.ipynb ├── reward_model_Bradley_Terry_model.ipynb ├── reward_model_ppo.ipynb ├── sft_llama2.ipynb └── trl_reward_trainer.ipynb ├── sentencepiece.ipynb ├── supplement_01_svd.ipynb ├── tokenizer.model ├── tokenizer ├── EsperBERTo │ ├── merges.txt │ └── vocab.json ├── how_to_train_a_tokenizer.ipynb ├── pad_eos.ipynb ├── tokenizer_gpt4o.ipynb ├── tokenizer_supply.ipynb └── xyj.txt ├── tokenizer_basics.ipynb ├── tokenizer_与词表拓展.ipynb ├── transformers_相关包及安装.ipynb ├── tricks ├── Speculative_Decoding.ipynb ├── multi-task-learning.ipynb └── speculative_decoding.png ├── trl_basics.ipynb ├── trl_hf ├── DPO_Final.pdf ├── dpo_math.ipynb ├── imgs │ ├── instructGPT.jpeg │ ├── lm_rl.png │ └── rlhf_dpo.png ├── lm_meets_rl.ipynb ├── rlhf_ppo_vs_dpo.ipynb ├── trl_basics.ipynb ├── trl_dpo.ipynb ├── trl_ppo_basics.ipynb ├── trl_ppo_pipeline.ipynb ├── trl_ppo_step.ipynb ├── trl_ppotrainer.ipynb ├── trl_ppotrainer_helloworld.ipynb └── trl_trainer_utils.ipynb ├── 数据集 └── 经典-标准数据集.ipynb └── 领域大模型 └── 医疗(medical).ipynb /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/.gitignore -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/README.md -------------------------------------------------------------------------------- /Stanford_cs229.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/Stanford_cs229.ipynb -------------------------------------------------------------------------------- /illya_NeurIPS_2024.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/illya_NeurIPS_2024.ipynb -------------------------------------------------------------------------------- /imgs/2880px-MCTS_(English).svg.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/2880px-MCTS_(English).svg.png -------------------------------------------------------------------------------- /imgs/4d_rotation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/4d_rotation.png -------------------------------------------------------------------------------- /imgs/accelerate.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/accelerate.png -------------------------------------------------------------------------------- /imgs/adap_kl_coef.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/adap_kl_coef.png -------------------------------------------------------------------------------- /imgs/agent_rl.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/agent_rl.png -------------------------------------------------------------------------------- /imgs/alpaca-instruct.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/alpaca-instruct.png -------------------------------------------------------------------------------- /imgs/bf-float.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/bf-float.png -------------------------------------------------------------------------------- /imgs/bloom.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/bloom.png -------------------------------------------------------------------------------- /imgs/bloom_lora.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/bloom_lora.png -------------------------------------------------------------------------------- /imgs/bpe.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/bpe.png -------------------------------------------------------------------------------- /imgs/c4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/c4.png -------------------------------------------------------------------------------- /imgs/cg.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/cg.png -------------------------------------------------------------------------------- /imgs/ch-dataset.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/ch-dataset.png -------------------------------------------------------------------------------- /imgs/cope.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/cope.png -------------------------------------------------------------------------------- /imgs/decoding-qkv.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/decoding-qkv.png -------------------------------------------------------------------------------- /imgs/drl.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/drl.png -------------------------------------------------------------------------------- /imgs/en-dataset.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/en-dataset.png -------------------------------------------------------------------------------- /imgs/float16.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/float16.png -------------------------------------------------------------------------------- /imgs/floats-rep.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/floats-rep.png -------------------------------------------------------------------------------- /imgs/generating.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/generating.png -------------------------------------------------------------------------------- /imgs/gpt-2-autoregression-2.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/gpt-2-autoregression-2.gif -------------------------------------------------------------------------------- /imgs/gpt2_bert_training.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/gpt2_bert_training.png -------------------------------------------------------------------------------- /imgs/gpt4-moe.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/gpt4-moe.png -------------------------------------------------------------------------------- /imgs/gpt4-price.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/gpt4-price.png -------------------------------------------------------------------------------- /imgs/gpt_config.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/gpt_config.jpeg -------------------------------------------------------------------------------- /imgs/gqa.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/gqa.png -------------------------------------------------------------------------------- /imgs/grad-accumulation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/grad-accumulation.png -------------------------------------------------------------------------------- /imgs/hf_llama_rope.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/hf_llama_rope.png -------------------------------------------------------------------------------- /imgs/kv-cache.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/kv-cache.gif -------------------------------------------------------------------------------- /imgs/kvcache-arch.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/kvcache-arch.png -------------------------------------------------------------------------------- /imgs/kvmatrices.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/kvmatrices.png -------------------------------------------------------------------------------- /imgs/llama2-weights.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/llama2-weights.png -------------------------------------------------------------------------------- /imgs/llama_weights.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/llama_weights.png -------------------------------------------------------------------------------- /imgs/llm-archs.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/llm-archs.png -------------------------------------------------------------------------------- /imgs/llm-int8.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/llm-int8.png -------------------------------------------------------------------------------- /imgs/llm-tokenizer.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/llm-tokenizer.png -------------------------------------------------------------------------------- /imgs/lora-weights.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/lora-weights.png -------------------------------------------------------------------------------- /imgs/lora.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/lora.png -------------------------------------------------------------------------------- /imgs/meta-llama-model.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/meta-llama-model.png -------------------------------------------------------------------------------- /imgs/model-base-model-free.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/model-base-model-free.png -------------------------------------------------------------------------------- /imgs/model_clm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/model_clm.png -------------------------------------------------------------------------------- /imgs/model_mlm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/model_mlm.png -------------------------------------------------------------------------------- /imgs/model_seq2seq.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/model_seq2seq.png -------------------------------------------------------------------------------- /imgs/moe.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/moe.png -------------------------------------------------------------------------------- /imgs/openrlhf.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/openrlhf.png -------------------------------------------------------------------------------- /imgs/peft.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/peft.png -------------------------------------------------------------------------------- /imgs/pile.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/pile.png -------------------------------------------------------------------------------- /imgs/ppo_algo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/ppo_algo.png -------------------------------------------------------------------------------- /imgs/qkv.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/qkv.png -------------------------------------------------------------------------------- /imgs/quant-freeze.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/quant-freeze.png -------------------------------------------------------------------------------- /imgs/regular-finetuning-alt.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/regular-finetuning-alt.png -------------------------------------------------------------------------------- /imgs/regular-finetuning.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/regular-finetuning.png -------------------------------------------------------------------------------- /imgs/rel_position_matrix.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/rel_position_matrix.png -------------------------------------------------------------------------------- /imgs/rlhf.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/rlhf.png -------------------------------------------------------------------------------- /imgs/rm_ppo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/rm_ppo.png -------------------------------------------------------------------------------- /imgs/rope_1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/rope_1.png -------------------------------------------------------------------------------- /imgs/rope_2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/rope_2.png -------------------------------------------------------------------------------- /imgs/rope_paper.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/rope_paper.png -------------------------------------------------------------------------------- /imgs/self-instruct.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/self-instruct.png -------------------------------------------------------------------------------- /imgs/t+1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/t+1.png -------------------------------------------------------------------------------- /imgs/token-kvcache.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/token-kvcache.png -------------------------------------------------------------------------------- /imgs/token_steps.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/token_steps.png -------------------------------------------------------------------------------- /imgs/trl-sft.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/trl-sft.png -------------------------------------------------------------------------------- /imgs/trl_overview.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/trl_overview.png -------------------------------------------------------------------------------- /imgs/trl_pipeline.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/trl_pipeline.png -------------------------------------------------------------------------------- /imgs/trl_ppo_loss.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/trl_ppo_loss.png -------------------------------------------------------------------------------- /imgs/trl_steps.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/trl_steps.png -------------------------------------------------------------------------------- /imgs/unique-tokens.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/unique-tokens.jpeg -------------------------------------------------------------------------------- /imgs/value_iteration.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/imgs/value_iteration.png -------------------------------------------------------------------------------- /tutorials/01_llama_intro.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/01_llama_intro.ipynb -------------------------------------------------------------------------------- /tutorials/02_lora_baics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/02_lora_baics.ipynb -------------------------------------------------------------------------------- /tutorials/03_lora_practice.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/03_lora_practice.ipynb -------------------------------------------------------------------------------- /tutorials/04_PEFT_lora.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/04_PEFT_lora.ipynb -------------------------------------------------------------------------------- /tutorials/05_float16_bf16.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/05_float16_bf16.ipynb -------------------------------------------------------------------------------- /tutorials/06_alpaca_llama_7b_lora.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/06_alpaca_llama_7b_lora.ipynb -------------------------------------------------------------------------------- /tutorials/07_fp16_mixed_precision_training.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/07_fp16_mixed_precision_training.ipynb -------------------------------------------------------------------------------- /tutorials/Chinese(中文)/chinese_llama2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/Chinese(中文)/chinese_llama2.ipynb -------------------------------------------------------------------------------- /tutorials/LLM.int8.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/LLM.int8.ipynb -------------------------------------------------------------------------------- /tutorials/LLMs & NLPs.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/LLMs & NLPs.ipynb -------------------------------------------------------------------------------- /tutorials/alpaca_vicuna.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/alpaca_vicuna.ipynb -------------------------------------------------------------------------------- /tutorials/attention/flash_attention.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/attention/flash_attention.ipynb -------------------------------------------------------------------------------- /tutorials/bitsandbytes.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/bitsandbytes.ipynb -------------------------------------------------------------------------------- /tutorials/bloom_源码分析.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/bloom_源码分析.ipynb -------------------------------------------------------------------------------- /tutorials/chatgpt_reviews.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/chatgpt_reviews.ipynb -------------------------------------------------------------------------------- /tutorials/code_sql/overall.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/code_sql/overall.ipynb -------------------------------------------------------------------------------- /tutorials/cuda_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/cuda_basics.ipynb -------------------------------------------------------------------------------- /tutorials/discussion/imgs/search_methods.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/discussion/imgs/search_methods.png -------------------------------------------------------------------------------- /tutorials/discussion/o1.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/discussion/o1.ipynb -------------------------------------------------------------------------------- /tutorials/drl/01_q_learning_dqn.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/01_q_learning_dqn.ipynb -------------------------------------------------------------------------------- /tutorials/drl/imgs/REINFOCE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/REINFOCE.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/ac_actor_pg.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/ac_actor_pg.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/ac_algo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/ac_algo.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/ac_critic_td.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/ac_critic_td.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/agent_as_nn.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/agent_as_nn.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/discounted_reward.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/discounted_reward.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/game_mcts_1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/game_mcts_1.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/game_mcts_2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/game_mcts_2.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/game_mcts_3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/game_mcts_3.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/game_mcts_4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/game_mcts_4.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/her_idea.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/her_idea.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/importance_sampling_issue.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/importance_sampling_issue.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/lm_rl.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/lm_rl.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/mab-definition.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/mab-definition.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/mab_algo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/mab_algo.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/mc_estimate_v_pi.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/mc_estimate_v_pi.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/mcts_algo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/mcts_algo.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/mcts_exam_1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/mcts_exam_1.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/mcts_exam_2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/mcts_exam_2.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/mcts_exam_3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/mcts_exam_3.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/mcts_exam_4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/mcts_exam_4.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/node-edge.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/node-edge.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/pg_gradient.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/pg_gradient.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/pg_is_advantage.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/pg_is_advantage.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/pg_loss.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/pg_loss.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/pieter_abbeel_trpo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/pieter_abbeel_trpo.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/policy_net_update.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/policy_net_update.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/policy_net_value_net_sharing.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/policy_net_value_net_sharing.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/policy_value_update_summary.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/policy_value_update_summary.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/ppo2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/ppo2.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/ppo_algo_lhy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/ppo_algo_lhy.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/ppo_clip.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/ppo_clip.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/ppo_clip_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/ppo_clip_curve.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/q_values.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/q_values.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/reinforce_monte_carlo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/reinforce_monte_carlo.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/reinforce_with_baseline_approx.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/reinforce_with_baseline_approx.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/rl_setting.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/rl_setting.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/rollout.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/rollout.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/s_a_r.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/s_a_r.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/s_a_r_p.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/s_a_r_p.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/s_a_r_pi.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/s_a_r_pi.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/suitable_credit.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/suitable_credit.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/td_learning.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/td_learning.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/td_trip.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/td_trip.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/td_update.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/td_update.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/td_v_pi.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/td_v_pi.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/trpo_Lagrange_duality.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/trpo_Lagrange_duality.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/trpo_obj.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/trpo_obj.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/trpo_summary.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/trpo_summary.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/value_iter.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/value_iter.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/value_loss.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/value_loss.png -------------------------------------------------------------------------------- /tutorials/drl/imgs/value_net_update.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/imgs/value_net_update.png -------------------------------------------------------------------------------- /tutorials/drl/mcts/alpha_zero.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/alpha_zero.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mcts/alphago_mcts.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/alphago_mcts.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mcts/mcts_01_intro_bascis.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/mcts_01_intro_bascis.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mcts/mcts_02_from_scartch.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/mcts_02_from_scartch.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mcts/mcts_03_binary_tree.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/mcts_03_binary_tree.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mcts/mcts_04_rl.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/mcts_04_rl.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mcts/mcts_05_tic-tac-toe.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/mcts_05_tic-tac-toe.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mcts/mcts_06_vis.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/mcts_06_vis.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mcts/mcts_demo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/mcts_demo.py -------------------------------------------------------------------------------- /tutorials/drl/mcts/mcts_supp.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mcts/mcts_supp.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mdp_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mdp_basics.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mdps_policy_gradient.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mdps_policy_gradient.ipynb -------------------------------------------------------------------------------- /tutorials/drl/mdps_value_iteration.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/mdps_value_iteration.ipynb -------------------------------------------------------------------------------- /tutorials/drl/multi_armed_bandits.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/multi_armed_bandits.ipynb -------------------------------------------------------------------------------- /tutorials/drl/o1_self_rl.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/o1_self_rl.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_based/PPO_lihongyi_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_based/PPO_lihongyi_basics.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_based/actor_critic.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_based/actor_critic.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_based/annotated_trpo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_based/annotated_trpo.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_based/pg_trpo_ppo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_based/pg_trpo_ppo.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_based/ppo_gae.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_based/ppo_gae.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_based/ppo_supp.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_based/ppo_supp.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_based/trpo_optimization.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_based/trpo_optimization.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_based_rl.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_based_rl.ipynb -------------------------------------------------------------------------------- /tutorials/drl/policy_gradient_with_basline.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/policy_gradient_with_basline.ipynb -------------------------------------------------------------------------------- /tutorials/drl/pomdp.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/pomdp.ipynb -------------------------------------------------------------------------------- /tutorials/drl/rl_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/rl_basics.ipynb -------------------------------------------------------------------------------- /tutorials/drl/rl_intro_review.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/rl_intro_review.ipynb -------------------------------------------------------------------------------- /tutorials/drl/rlhf/RM_loss.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/rlhf/RM_loss.ipynb -------------------------------------------------------------------------------- /tutorials/drl/rl应用/供应链优化.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/rl应用/供应链优化.ipynb -------------------------------------------------------------------------------- /tutorials/drl/temporal_difference_bootstrapping.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/temporal_difference_bootstrapping.ipynb -------------------------------------------------------------------------------- /tutorials/drl/utils/gridworld.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/utils/gridworld.py -------------------------------------------------------------------------------- /tutorials/drl/utils/mdp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/utils/mdp.py -------------------------------------------------------------------------------- /tutorials/drl/utils/rendering_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/utils/rendering_utils.py -------------------------------------------------------------------------------- /tutorials/drl/utils/test.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/utils/test.ipynb -------------------------------------------------------------------------------- /tutorials/drl/value_based/q_learning_lihongyi.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/value_based/q_learning_lihongyi.ipynb -------------------------------------------------------------------------------- /tutorials/drl/value_based/td.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/value_based/td.ipynb -------------------------------------------------------------------------------- /tutorials/drl/对(深度)强化学习的认识与理解.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/对(深度)强化学习的认识与理解.ipynb -------------------------------------------------------------------------------- /tutorials/drl/经典 envs.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/drl/经典 envs.ipynb -------------------------------------------------------------------------------- /tutorials/evaluation/datasets与大模型评估(evaluation).ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/evaluation/datasets与大模型评估(evaluation).ipynb -------------------------------------------------------------------------------- /tutorials/evaluation/test_questions.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/evaluation/test_questions.ipynb -------------------------------------------------------------------------------- /tutorials/falcon.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/falcon.ipynb -------------------------------------------------------------------------------- /tutorials/gpt2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/gpt2.ipynb -------------------------------------------------------------------------------- /tutorials/gpt2_train_tokenizer.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/gpt2_train_tokenizer.ipynb -------------------------------------------------------------------------------- /tutorials/gpt_family.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/gpt_family.ipynb -------------------------------------------------------------------------------- /tutorials/gradient_accumulation.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/gradient_accumulation.ipynb -------------------------------------------------------------------------------- /tutorials/gradient_checkpointing.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/gradient_checkpointing.ipynb -------------------------------------------------------------------------------- /tutorials/instruction_chat/chat_template.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/instruction_chat/chat_template.ipynb -------------------------------------------------------------------------------- /tutorials/karpathy/01_build_GPT.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/karpathy/01_build_GPT.ipynb -------------------------------------------------------------------------------- /tutorials/karpathy/input.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/karpathy/input.txt -------------------------------------------------------------------------------- /tutorials/llama/RewardModel.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama/RewardModel.ipynb -------------------------------------------------------------------------------- /tutorials/llama/attention_map_vis.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama/attention_map_vis.ipynb -------------------------------------------------------------------------------- /tutorials/llama/llama2_7b_llama3_8b.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama/llama2_7b_llama3_8b.ipynb -------------------------------------------------------------------------------- /tutorials/llama/llama2_tokenizer.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama/llama2_tokenizer.model -------------------------------------------------------------------------------- /tutorials/llama/llama3_0419.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama/llama3_0419.ipynb -------------------------------------------------------------------------------- /tutorials/llama/llama3_supp_data_config.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama/llama3_supp_data_config.ipynb -------------------------------------------------------------------------------- /tutorials/llama/llama3_tokenizer.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama/llama3_tokenizer.model -------------------------------------------------------------------------------- /tutorials/llama/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama/tokenizer.py -------------------------------------------------------------------------------- /tutorials/llama2_introduction.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_introduction.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_01.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_01.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_02_RMSNorm_SwiGLU.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_02_RMSNorm_SwiGLU.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_03_RoPE.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_03_RoPE.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_03_RoPE_01_complex_position_encoding.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_03_RoPE_01_complex_position_encoding.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_03_RoPE_apply_rotary_emb.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_03_RoPE_apply_rotary_emb.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_cache_kv.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_cache_kv.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_generate.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_generate.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_grouped_query_attention.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_grouped_query_attention.ipynb -------------------------------------------------------------------------------- /tutorials/llama2_src_text_chat.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama2_src_text_chat.ipynb -------------------------------------------------------------------------------- /tutorials/llama_tokenizer.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llama_tokenizer.ipynb -------------------------------------------------------------------------------- /tutorials/llm_training_sft.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llm_training_sft.ipynb -------------------------------------------------------------------------------- /tutorials/llms_models_datasets_finetune.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/llms_models_datasets_finetune.ipynb -------------------------------------------------------------------------------- /tutorials/lora_all_parameters.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/lora_all_parameters.ipynb -------------------------------------------------------------------------------- /tutorials/misc/gradio.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/misc/gradio.ipynb -------------------------------------------------------------------------------- /tutorials/model_finetune_quantization.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/model_finetune_quantization.ipynb -------------------------------------------------------------------------------- /tutorials/moe/Mistral_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/moe/Mistral_basics.ipynb -------------------------------------------------------------------------------- /tutorials/moe/basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/moe/basics.ipynb -------------------------------------------------------------------------------- /tutorials/moe/mistral_src_code.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/moe/mistral_src_code.ipynb -------------------------------------------------------------------------------- /tutorials/multi_model_vit_clip.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/multi_model_vit_clip.ipynb -------------------------------------------------------------------------------- /tutorials/peft/04_PEFT_lora.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/peft/04_PEFT_lora.ipynb -------------------------------------------------------------------------------- /tutorials/peft/PEFT_supp.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/peft/PEFT_supp.ipynb -------------------------------------------------------------------------------- /tutorials/peft/lora_inject.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/peft/lora_inject.ipynb -------------------------------------------------------------------------------- /tutorials/peft/lora_merge.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/peft/lora_merge.ipynb -------------------------------------------------------------------------------- /tutorials/peft/lora_svd_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/peft/lora_svd_basics.ipynb -------------------------------------------------------------------------------- /tutorials/peft/lora_vs_whole_model_1.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/peft/lora_vs_whole_model_1.ipynb -------------------------------------------------------------------------------- /tutorials/peft/lora_vs_whole_model_2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/peft/lora_vs_whole_model_2.ipynb -------------------------------------------------------------------------------- /tutorials/position_encoding.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/position_encoding.ipynb -------------------------------------------------------------------------------- /tutorials/position_encoding/long_context_window.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/position_encoding/long_context_window.ipynb -------------------------------------------------------------------------------- /tutorials/position_encoding/rope-beta-encoding.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/position_encoding/rope-beta-encoding.ipynb -------------------------------------------------------------------------------- /tutorials/position_encoding/rope_cope.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/position_encoding/rope_cope.ipynb -------------------------------------------------------------------------------- /tutorials/pretrain_corpus_大模型预训练语料.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/pretrain_corpus_大模型预训练语料.ipynb -------------------------------------------------------------------------------- /tutorials/quantization_gptq.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/quantization_gptq.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/Mistral-7b-dpo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/Mistral-7b-dpo.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/bradley_terry_modeling.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/bradley_terry_modeling.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/dpo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/dpo.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/dpo_llama2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/dpo_llama2.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/imgs/quality-labels.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/imgs/quality-labels.png -------------------------------------------------------------------------------- /tutorials/reward_model/ppo_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/ppo_basics.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/reward_model_Bradley_Terry_model.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/reward_model_Bradley_Terry_model.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/reward_model_ppo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/reward_model_ppo.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/sft_llama2.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/sft_llama2.ipynb -------------------------------------------------------------------------------- /tutorials/reward_model/trl_reward_trainer.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/reward_model/trl_reward_trainer.ipynb -------------------------------------------------------------------------------- /tutorials/sentencepiece.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/sentencepiece.ipynb -------------------------------------------------------------------------------- /tutorials/supplement_01_svd.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/supplement_01_svd.ipynb -------------------------------------------------------------------------------- /tutorials/tokenizer.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer.model -------------------------------------------------------------------------------- /tutorials/tokenizer/EsperBERTo/merges.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer/EsperBERTo/merges.txt -------------------------------------------------------------------------------- /tutorials/tokenizer/EsperBERTo/vocab.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer/EsperBERTo/vocab.json -------------------------------------------------------------------------------- /tutorials/tokenizer/how_to_train_a_tokenizer.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer/how_to_train_a_tokenizer.ipynb -------------------------------------------------------------------------------- /tutorials/tokenizer/pad_eos.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer/pad_eos.ipynb -------------------------------------------------------------------------------- /tutorials/tokenizer/tokenizer_gpt4o.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer/tokenizer_gpt4o.ipynb -------------------------------------------------------------------------------- /tutorials/tokenizer/tokenizer_supply.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer/tokenizer_supply.ipynb -------------------------------------------------------------------------------- /tutorials/tokenizer/xyj.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer/xyj.txt -------------------------------------------------------------------------------- /tutorials/tokenizer_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer_basics.ipynb -------------------------------------------------------------------------------- /tutorials/tokenizer_与词表拓展.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tokenizer_与词表拓展.ipynb -------------------------------------------------------------------------------- /tutorials/transformers_相关包及安装.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/transformers_相关包及安装.ipynb -------------------------------------------------------------------------------- /tutorials/tricks/Speculative_Decoding.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tricks/Speculative_Decoding.ipynb -------------------------------------------------------------------------------- /tutorials/tricks/multi-task-learning.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tricks/multi-task-learning.ipynb -------------------------------------------------------------------------------- /tutorials/tricks/speculative_decoding.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/tricks/speculative_decoding.png -------------------------------------------------------------------------------- /tutorials/trl_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_basics.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/DPO_Final.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/DPO_Final.pdf -------------------------------------------------------------------------------- /tutorials/trl_hf/dpo_math.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/dpo_math.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/imgs/instructGPT.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/imgs/instructGPT.jpeg -------------------------------------------------------------------------------- /tutorials/trl_hf/imgs/lm_rl.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/imgs/lm_rl.png -------------------------------------------------------------------------------- /tutorials/trl_hf/imgs/rlhf_dpo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/imgs/rlhf_dpo.png -------------------------------------------------------------------------------- /tutorials/trl_hf/lm_meets_rl.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/lm_meets_rl.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/rlhf_ppo_vs_dpo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/rlhf_ppo_vs_dpo.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/trl_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/trl_basics.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/trl_dpo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/trl_dpo.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/trl_ppo_basics.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/trl_ppo_basics.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/trl_ppo_pipeline.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/trl_ppo_pipeline.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/trl_ppo_step.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/trl_ppo_step.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/trl_ppotrainer.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/trl_ppotrainer.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/trl_ppotrainer_helloworld.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/trl_ppotrainer_helloworld.ipynb -------------------------------------------------------------------------------- /tutorials/trl_hf/trl_trainer_utils.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/trl_hf/trl_trainer_utils.ipynb -------------------------------------------------------------------------------- /tutorials/数据集/经典-标准数据集.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/数据集/经典-标准数据集.ipynb -------------------------------------------------------------------------------- /tutorials/领域大模型/医疗(medical).ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/chunhuizhang/personal_chatgpt/HEAD/tutorials/领域大模型/医疗(medical).ipynb --------------------------------------------------------------------------------