initial
Browse files- README.md +61 -0
- added_tokens.json +5 -0
- checkpoint-22500/added_tokens.json +5 -0
- checkpoint-22500/config.json +39 -0
- checkpoint-22500/generation_config.json +7 -0
- checkpoint-22500/merges.txt +0 -0
- checkpoint-22500/optimizer.pt +3 -0
- checkpoint-22500/pytorch_model.bin +3 -0
- checkpoint-22500/rng_state.pth +3 -0
- checkpoint-22500/scheduler.pt +3 -0
- checkpoint-22500/special_tokens_map.json +11 -0
- checkpoint-22500/tokenizer_config.json +35 -0
- checkpoint-22500/trainer_state.json +310 -0
- checkpoint-22500/training_args.bin +3 -0
- checkpoint-22500/vocab.json +0 -0
- checkpoint-30000/added_tokens.json +5 -0
- checkpoint-30000/config.json +39 -0
- checkpoint-30000/generation_config.json +7 -0
- checkpoint-30000/merges.txt +0 -0
- checkpoint-30000/optimizer.pt +3 -0
- checkpoint-30000/pytorch_model.bin +3 -0
- checkpoint-30000/rng_state.pth +3 -0
- checkpoint-30000/scheduler.pt +3 -0
- checkpoint-30000/special_tokens_map.json +11 -0
- checkpoint-30000/tokenizer_config.json +35 -0
- checkpoint-30000/trainer_state.json +408 -0
- checkpoint-30000/training_args.bin +3 -0
- checkpoint-30000/vocab.json +0 -0
- logs/events.out.tfevents.1690017287.7cfadc3aec44.25659.0 +3 -0
- logs/events.out.tfevents.1690017422.7cfadc3aec44.25659.1 +3 -0
- logs/events.out.tfevents.1690017606.7cfadc3aec44.27769.0 +3 -0
- logs/events.out.tfevents.1690017757.7cfadc3aec44.28627.0 +3 -0
- logs/events.out.tfevents.1690021418.7cfadc3aec44.28627.1 +3 -0
- merges.txt +0 -0
- special_tokens_map.json +11 -0
- tokenizer_config.json +35 -0
- vocab.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
base_model: vicgalle/gpt2-open-instruct-v1
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: hh-rlhf
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# hh-rlhf
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [vicgalle/gpt2-open-instruct-v1](https://huggingface.co/vicgalle/gpt2-open-instruct-v1) on an unknown dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 2.1534
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 0.0005
|
| 38 |
+
- train_batch_size: 2
|
| 39 |
+
- eval_batch_size: 1
|
| 40 |
+
- seed: 42
|
| 41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
+
- lr_scheduler_type: linear
|
| 43 |
+
- lr_scheduler_warmup_steps: 100
|
| 44 |
+
- num_epochs: 4
|
| 45 |
+
|
| 46 |
+
### Training results
|
| 47 |
+
|
| 48 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 49 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 50 |
+
| 2.3108 | 1.0 | 7500 | 2.1799 |
|
| 51 |
+
| 2.265 | 2.0 | 15000 | 2.1632 |
|
| 52 |
+
| 2.2507 | 3.0 | 22500 | 2.1567 |
|
| 53 |
+
| 2.2519 | 4.0 | 30000 | 2.1534 |
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
### Framework versions
|
| 57 |
+
|
| 58 |
+
- Transformers 4.31.0
|
| 59 |
+
- Pytorch 2.0.1+cu118
|
| 60 |
+
- Datasets 2.13.1
|
| 61 |
+
- Tokenizers 0.13.3
|
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"### End": 50257,
|
| 3 |
+
"### Instruction:": 50258,
|
| 4 |
+
"### Response:\n": 50259
|
| 5 |
+
}
|
checkpoint-22500/added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"### End": 50257,
|
| 3 |
+
"### Instruction:": 50258,
|
| 4 |
+
"### Response:\n": 50259
|
| 5 |
+
}
|
checkpoint-22500/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "vicgalle/gpt2-open-instruct-v1",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"reorder_and_upcast_attn": false,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.31.0",
|
| 37 |
+
"use_cache": false,
|
| 38 |
+
"vocab_size": 50260
|
| 39 |
+
}
|
checkpoint-22500/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.31.0",
|
| 6 |
+
"use_cache": false
|
| 7 |
+
}
|
checkpoint-22500/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-22500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bea68af7cf27f7c73c7fc2d1cb8e9481afb1f6c1cc4ef60d5d10ddd12c1909c0
|
| 3 |
+
size 2291
|
checkpoint-22500/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a16b875c379cf964f25eb0ad6dec0fd71176a6040583d2275df7fabab3a914c
|
| 3 |
+
size 497816413
|
checkpoint-22500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7bea4e9033f1d4e0414230b3e4247dbd6a4acc6839c67618dee70f08a3a9802c
|
| 3 |
+
size 14575
|
checkpoint-22500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d26001e27eef77119f809f5bef50d877811ea7917430110cedcc16d7e22156aa
|
| 3 |
+
size 563
|
checkpoint-22500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"### End",
|
| 4 |
+
"### Instruction:",
|
| 5 |
+
"### Response:\n"
|
| 6 |
+
],
|
| 7 |
+
"bos_token": "<|endoftext|>",
|
| 8 |
+
"eos_token": "<|endoftext|>",
|
| 9 |
+
"pad_token": "<|endoftext|>",
|
| 10 |
+
"unk_token": "<|endoftext|>"
|
| 11 |
+
}
|
checkpoint-22500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"bos_token": {
|
| 5 |
+
"__type": "AddedToken",
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"clean_up_tokenization_spaces": true,
|
| 13 |
+
"eos_token": {
|
| 14 |
+
"__type": "AddedToken",
|
| 15 |
+
"content": "<|endoftext|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": true,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false
|
| 20 |
+
},
|
| 21 |
+
"errors": "replace",
|
| 22 |
+
"max_seq_length": 512,
|
| 23 |
+
"model_max_length": 1024,
|
| 24 |
+
"pad_token": null,
|
| 25 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 26 |
+
"unk_token": {
|
| 27 |
+
"__type": "AddedToken",
|
| 28 |
+
"content": "<|endoftext|>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": true,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false
|
| 33 |
+
},
|
| 34 |
+
"use_fast": false
|
| 35 |
+
}
|
checkpoint-22500/trainer_state.json
ADDED
|
@@ -0,0 +1,310 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.156663179397583,
|
| 3 |
+
"best_model_checkpoint": "gpt2-open-instruct-v1-Anthropic/hh-rlhf/checkpoint-22500",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 22500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.07,
|
| 12 |
+
"learning_rate": 0.009867224080267559,
|
| 13 |
+
"loss": 2.4868,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.13,
|
| 18 |
+
"learning_rate": 0.0097,
|
| 19 |
+
"loss": 2.3663,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.2,
|
| 24 |
+
"learning_rate": 0.009532775919732442,
|
| 25 |
+
"loss": 2.3524,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.27,
|
| 30 |
+
"learning_rate": 0.009365551839464883,
|
| 31 |
+
"loss": 2.3374,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.33,
|
| 36 |
+
"learning_rate": 0.009198327759197324,
|
| 37 |
+
"loss": 2.3286,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.4,
|
| 42 |
+
"learning_rate": 0.009031103678929767,
|
| 43 |
+
"loss": 2.3082,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.47,
|
| 48 |
+
"learning_rate": 0.008863879598662208,
|
| 49 |
+
"loss": 2.3161,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.53,
|
| 54 |
+
"learning_rate": 0.00869665551839465,
|
| 55 |
+
"loss": 2.308,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.6,
|
| 60 |
+
"learning_rate": 0.00852943143812709,
|
| 61 |
+
"loss": 2.3175,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.67,
|
| 66 |
+
"learning_rate": 0.008362541806020068,
|
| 67 |
+
"loss": 2.3023,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.73,
|
| 72 |
+
"learning_rate": 0.008195317725752507,
|
| 73 |
+
"loss": 2.3155,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.8,
|
| 78 |
+
"learning_rate": 0.00802809364548495,
|
| 79 |
+
"loss": 2.314,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.87,
|
| 84 |
+
"learning_rate": 0.007860869565217391,
|
| 85 |
+
"loss": 2.2981,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.93,
|
| 90 |
+
"learning_rate": 0.007693645484949833,
|
| 91 |
+
"loss": 2.3153,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.0,
|
| 96 |
+
"learning_rate": 0.007526755852842809,
|
| 97 |
+
"loss": 2.3108,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0,
|
| 102 |
+
"eval_loss": 2.1798622608184814,
|
| 103 |
+
"eval_runtime": 12.4843,
|
| 104 |
+
"eval_samples_per_second": 40.05,
|
| 105 |
+
"eval_steps_per_second": 40.05,
|
| 106 |
+
"step": 7500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.07,
|
| 110 |
+
"learning_rate": 0.007359531772575251,
|
| 111 |
+
"loss": 2.2738,
|
| 112 |
+
"step": 8000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.13,
|
| 116 |
+
"learning_rate": 0.007192307692307692,
|
| 117 |
+
"loss": 2.2885,
|
| 118 |
+
"step": 8500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.2,
|
| 122 |
+
"learning_rate": 0.007025083612040134,
|
| 123 |
+
"loss": 2.313,
|
| 124 |
+
"step": 9000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.27,
|
| 128 |
+
"learning_rate": 0.006858193979933111,
|
| 129 |
+
"loss": 2.2854,
|
| 130 |
+
"step": 9500
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.33,
|
| 134 |
+
"learning_rate": 0.006690969899665552,
|
| 135 |
+
"loss": 2.2524,
|
| 136 |
+
"step": 10000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.4,
|
| 140 |
+
"learning_rate": 0.006523745819397994,
|
| 141 |
+
"loss": 2.2767,
|
| 142 |
+
"step": 10500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.47,
|
| 146 |
+
"learning_rate": 0.006356521739130435,
|
| 147 |
+
"loss": 2.2627,
|
| 148 |
+
"step": 11000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.53,
|
| 152 |
+
"learning_rate": 0.006189297658862877,
|
| 153 |
+
"loss": 2.2779,
|
| 154 |
+
"step": 11500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.6,
|
| 158 |
+
"learning_rate": 0.0060224080267558535,
|
| 159 |
+
"loss": 2.2778,
|
| 160 |
+
"step": 12000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.67,
|
| 164 |
+
"learning_rate": 0.005855183946488294,
|
| 165 |
+
"loss": 2.2588,
|
| 166 |
+
"step": 12500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.73,
|
| 170 |
+
"learning_rate": 0.005687959866220735,
|
| 171 |
+
"loss": 2.296,
|
| 172 |
+
"step": 13000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.8,
|
| 176 |
+
"learning_rate": 0.005520735785953177,
|
| 177 |
+
"loss": 2.2451,
|
| 178 |
+
"step": 13500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.87,
|
| 182 |
+
"learning_rate": 0.0053538461538461535,
|
| 183 |
+
"loss": 2.2831,
|
| 184 |
+
"step": 14000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.93,
|
| 188 |
+
"learning_rate": 0.005186622073578596,
|
| 189 |
+
"loss": 2.2801,
|
| 190 |
+
"step": 14500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 2.0,
|
| 194 |
+
"learning_rate": 0.005019397993311037,
|
| 195 |
+
"loss": 2.265,
|
| 196 |
+
"step": 15000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.0,
|
| 200 |
+
"eval_loss": 2.1632304191589355,
|
| 201 |
+
"eval_runtime": 12.092,
|
| 202 |
+
"eval_samples_per_second": 41.35,
|
| 203 |
+
"eval_steps_per_second": 41.35,
|
| 204 |
+
"step": 15000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.07,
|
| 208 |
+
"learning_rate": 0.004852173913043478,
|
| 209 |
+
"loss": 2.2483,
|
| 210 |
+
"step": 15500
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.13,
|
| 214 |
+
"learning_rate": 0.00468494983277592,
|
| 215 |
+
"loss": 2.2928,
|
| 216 |
+
"step": 16000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.2,
|
| 220 |
+
"learning_rate": 0.004518060200668896,
|
| 221 |
+
"loss": 2.2555,
|
| 222 |
+
"step": 16500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.27,
|
| 226 |
+
"learning_rate": 0.004350836120401338,
|
| 227 |
+
"loss": 2.2761,
|
| 228 |
+
"step": 17000
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.33,
|
| 232 |
+
"learning_rate": 0.00418361204013378,
|
| 233 |
+
"loss": 2.2515,
|
| 234 |
+
"step": 17500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.4,
|
| 238 |
+
"learning_rate": 0.004016387959866221,
|
| 239 |
+
"loss": 2.2636,
|
| 240 |
+
"step": 18000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.47,
|
| 244 |
+
"learning_rate": 0.003849498327759197,
|
| 245 |
+
"loss": 2.2569,
|
| 246 |
+
"step": 18500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.53,
|
| 250 |
+
"learning_rate": 0.003682274247491639,
|
| 251 |
+
"loss": 2.2728,
|
| 252 |
+
"step": 19000
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.6,
|
| 256 |
+
"learning_rate": 0.0035150501672240805,
|
| 257 |
+
"loss": 2.2561,
|
| 258 |
+
"step": 19500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.67,
|
| 262 |
+
"learning_rate": 0.0033478260869565218,
|
| 263 |
+
"loss": 2.2396,
|
| 264 |
+
"step": 20000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.73,
|
| 268 |
+
"learning_rate": 0.0031809364548494984,
|
| 269 |
+
"loss": 2.2601,
|
| 270 |
+
"step": 20500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.8,
|
| 274 |
+
"learning_rate": 0.00301371237458194,
|
| 275 |
+
"loss": 2.2609,
|
| 276 |
+
"step": 21000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 2.87,
|
| 280 |
+
"learning_rate": 0.002846488294314381,
|
| 281 |
+
"loss": 2.2654,
|
| 282 |
+
"step": 21500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 2.93,
|
| 286 |
+
"learning_rate": 0.0026792642140468226,
|
| 287 |
+
"loss": 2.2381,
|
| 288 |
+
"step": 22000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.0,
|
| 292 |
+
"learning_rate": 0.0025123745819397992,
|
| 293 |
+
"loss": 2.2507,
|
| 294 |
+
"step": 22500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.0,
|
| 298 |
+
"eval_loss": 2.156663179397583,
|
| 299 |
+
"eval_runtime": 12.0987,
|
| 300 |
+
"eval_samples_per_second": 41.327,
|
| 301 |
+
"eval_steps_per_second": 41.327,
|
| 302 |
+
"step": 22500
|
| 303 |
+
}
|
| 304 |
+
],
|
| 305 |
+
"max_steps": 30000,
|
| 306 |
+
"num_train_epochs": 4,
|
| 307 |
+
"total_flos": 2.351628288e+16,
|
| 308 |
+
"trial_name": null,
|
| 309 |
+
"trial_params": null
|
| 310 |
+
}
|
checkpoint-22500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b266296d18d0d931e280725174f6a350cf988e8a9bc67cb4310bffb7b37b68e4
|
| 3 |
+
size 3963
|
checkpoint-22500/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-30000/added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"### End": 50257,
|
| 3 |
+
"### Instruction:": 50258,
|
| 4 |
+
"### Response:\n": 50259
|
| 5 |
+
}
|
checkpoint-30000/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "vicgalle/gpt2-open-instruct-v1",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"reorder_and_upcast_attn": false,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.31.0",
|
| 37 |
+
"use_cache": false,
|
| 38 |
+
"vocab_size": 50260
|
| 39 |
+
}
|
checkpoint-30000/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.31.0",
|
| 6 |
+
"use_cache": false
|
| 7 |
+
}
|
checkpoint-30000/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-30000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64b4f2039c217bae5cf490dbee539ce0055a1d718b070f82d1ebf7f35d80a7b1
|
| 3 |
+
size 2291
|
checkpoint-30000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2da526274d343907df600f54a509d30a6454713992637363590ec30d915b8f8
|
| 3 |
+
size 497816413
|
checkpoint-30000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b70b701a5e1a011f3c2d46e48dd9aff4ec6351636d3447e8831c5dcdc0e79ab4
|
| 3 |
+
size 14575
|
checkpoint-30000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c803a4ee3bd255514329251efc3d78e44886b6920a7415f394699b3d7e88e542
|
| 3 |
+
size 563
|
checkpoint-30000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"### End",
|
| 4 |
+
"### Instruction:",
|
| 5 |
+
"### Response:\n"
|
| 6 |
+
],
|
| 7 |
+
"bos_token": "<|endoftext|>",
|
| 8 |
+
"eos_token": "<|endoftext|>",
|
| 9 |
+
"pad_token": "<|endoftext|>",
|
| 10 |
+
"unk_token": "<|endoftext|>"
|
| 11 |
+
}
|
checkpoint-30000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"bos_token": {
|
| 5 |
+
"__type": "AddedToken",
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"clean_up_tokenization_spaces": true,
|
| 13 |
+
"eos_token": {
|
| 14 |
+
"__type": "AddedToken",
|
| 15 |
+
"content": "<|endoftext|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": true,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false
|
| 20 |
+
},
|
| 21 |
+
"errors": "replace",
|
| 22 |
+
"max_seq_length": 512,
|
| 23 |
+
"model_max_length": 1024,
|
| 24 |
+
"pad_token": null,
|
| 25 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 26 |
+
"unk_token": {
|
| 27 |
+
"__type": "AddedToken",
|
| 28 |
+
"content": "<|endoftext|>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": true,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false
|
| 33 |
+
},
|
| 34 |
+
"use_fast": false
|
| 35 |
+
}
|
checkpoint-30000/trainer_state.json
ADDED
|
@@ -0,0 +1,408 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.1534345149993896,
|
| 3 |
+
"best_model_checkpoint": "gpt2-open-instruct-v1-Anthropic/hh-rlhf/checkpoint-30000",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 30000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.07,
|
| 12 |
+
"learning_rate": 0.009867224080267559,
|
| 13 |
+
"loss": 2.4868,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.13,
|
| 18 |
+
"learning_rate": 0.0097,
|
| 19 |
+
"loss": 2.3663,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.2,
|
| 24 |
+
"learning_rate": 0.009532775919732442,
|
| 25 |
+
"loss": 2.3524,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.27,
|
| 30 |
+
"learning_rate": 0.009365551839464883,
|
| 31 |
+
"loss": 2.3374,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.33,
|
| 36 |
+
"learning_rate": 0.009198327759197324,
|
| 37 |
+
"loss": 2.3286,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.4,
|
| 42 |
+
"learning_rate": 0.009031103678929767,
|
| 43 |
+
"loss": 2.3082,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.47,
|
| 48 |
+
"learning_rate": 0.008863879598662208,
|
| 49 |
+
"loss": 2.3161,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.53,
|
| 54 |
+
"learning_rate": 0.00869665551839465,
|
| 55 |
+
"loss": 2.308,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.6,
|
| 60 |
+
"learning_rate": 0.00852943143812709,
|
| 61 |
+
"loss": 2.3175,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.67,
|
| 66 |
+
"learning_rate": 0.008362541806020068,
|
| 67 |
+
"loss": 2.3023,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.73,
|
| 72 |
+
"learning_rate": 0.008195317725752507,
|
| 73 |
+
"loss": 2.3155,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.8,
|
| 78 |
+
"learning_rate": 0.00802809364548495,
|
| 79 |
+
"loss": 2.314,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.87,
|
| 84 |
+
"learning_rate": 0.007860869565217391,
|
| 85 |
+
"loss": 2.2981,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.93,
|
| 90 |
+
"learning_rate": 0.007693645484949833,
|
| 91 |
+
"loss": 2.3153,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.0,
|
| 96 |
+
"learning_rate": 0.007526755852842809,
|
| 97 |
+
"loss": 2.3108,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0,
|
| 102 |
+
"eval_loss": 2.1798622608184814,
|
| 103 |
+
"eval_runtime": 12.4843,
|
| 104 |
+
"eval_samples_per_second": 40.05,
|
| 105 |
+
"eval_steps_per_second": 40.05,
|
| 106 |
+
"step": 7500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.07,
|
| 110 |
+
"learning_rate": 0.007359531772575251,
|
| 111 |
+
"loss": 2.2738,
|
| 112 |
+
"step": 8000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.13,
|
| 116 |
+
"learning_rate": 0.007192307692307692,
|
| 117 |
+
"loss": 2.2885,
|
| 118 |
+
"step": 8500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.2,
|
| 122 |
+
"learning_rate": 0.007025083612040134,
|
| 123 |
+
"loss": 2.313,
|
| 124 |
+
"step": 9000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.27,
|
| 128 |
+
"learning_rate": 0.006858193979933111,
|
| 129 |
+
"loss": 2.2854,
|
| 130 |
+
"step": 9500
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.33,
|
| 134 |
+
"learning_rate": 0.006690969899665552,
|
| 135 |
+
"loss": 2.2524,
|
| 136 |
+
"step": 10000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.4,
|
| 140 |
+
"learning_rate": 0.006523745819397994,
|
| 141 |
+
"loss": 2.2767,
|
| 142 |
+
"step": 10500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.47,
|
| 146 |
+
"learning_rate": 0.006356521739130435,
|
| 147 |
+
"loss": 2.2627,
|
| 148 |
+
"step": 11000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.53,
|
| 152 |
+
"learning_rate": 0.006189297658862877,
|
| 153 |
+
"loss": 2.2779,
|
| 154 |
+
"step": 11500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.6,
|
| 158 |
+
"learning_rate": 0.0060224080267558535,
|
| 159 |
+
"loss": 2.2778,
|
| 160 |
+
"step": 12000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.67,
|
| 164 |
+
"learning_rate": 0.005855183946488294,
|
| 165 |
+
"loss": 2.2588,
|
| 166 |
+
"step": 12500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.73,
|
| 170 |
+
"learning_rate": 0.005687959866220735,
|
| 171 |
+
"loss": 2.296,
|
| 172 |
+
"step": 13000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.8,
|
| 176 |
+
"learning_rate": 0.005520735785953177,
|
| 177 |
+
"loss": 2.2451,
|
| 178 |
+
"step": 13500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.87,
|
| 182 |
+
"learning_rate": 0.0053538461538461535,
|
| 183 |
+
"loss": 2.2831,
|
| 184 |
+
"step": 14000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.93,
|
| 188 |
+
"learning_rate": 0.005186622073578596,
|
| 189 |
+
"loss": 2.2801,
|
| 190 |
+
"step": 14500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 2.0,
|
| 194 |
+
"learning_rate": 0.005019397993311037,
|
| 195 |
+
"loss": 2.265,
|
| 196 |
+
"step": 15000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.0,
|
| 200 |
+
"eval_loss": 2.1632304191589355,
|
| 201 |
+
"eval_runtime": 12.092,
|
| 202 |
+
"eval_samples_per_second": 41.35,
|
| 203 |
+
"eval_steps_per_second": 41.35,
|
| 204 |
+
"step": 15000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.07,
|
| 208 |
+
"learning_rate": 0.004852173913043478,
|
| 209 |
+
"loss": 2.2483,
|
| 210 |
+
"step": 15500
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.13,
|
| 214 |
+
"learning_rate": 0.00468494983277592,
|
| 215 |
+
"loss": 2.2928,
|
| 216 |
+
"step": 16000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.2,
|
| 220 |
+
"learning_rate": 0.004518060200668896,
|
| 221 |
+
"loss": 2.2555,
|
| 222 |
+
"step": 16500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.27,
|
| 226 |
+
"learning_rate": 0.004350836120401338,
|
| 227 |
+
"loss": 2.2761,
|
| 228 |
+
"step": 17000
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.33,
|
| 232 |
+
"learning_rate": 0.00418361204013378,
|
| 233 |
+
"loss": 2.2515,
|
| 234 |
+
"step": 17500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.4,
|
| 238 |
+
"learning_rate": 0.004016387959866221,
|
| 239 |
+
"loss": 2.2636,
|
| 240 |
+
"step": 18000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.47,
|
| 244 |
+
"learning_rate": 0.003849498327759197,
|
| 245 |
+
"loss": 2.2569,
|
| 246 |
+
"step": 18500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.53,
|
| 250 |
+
"learning_rate": 0.003682274247491639,
|
| 251 |
+
"loss": 2.2728,
|
| 252 |
+
"step": 19000
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.6,
|
| 256 |
+
"learning_rate": 0.0035150501672240805,
|
| 257 |
+
"loss": 2.2561,
|
| 258 |
+
"step": 19500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.67,
|
| 262 |
+
"learning_rate": 0.0033478260869565218,
|
| 263 |
+
"loss": 2.2396,
|
| 264 |
+
"step": 20000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.73,
|
| 268 |
+
"learning_rate": 0.0031809364548494984,
|
| 269 |
+
"loss": 2.2601,
|
| 270 |
+
"step": 20500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.8,
|
| 274 |
+
"learning_rate": 0.00301371237458194,
|
| 275 |
+
"loss": 2.2609,
|
| 276 |
+
"step": 21000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 2.87,
|
| 280 |
+
"learning_rate": 0.002846488294314381,
|
| 281 |
+
"loss": 2.2654,
|
| 282 |
+
"step": 21500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 2.93,
|
| 286 |
+
"learning_rate": 0.0026792642140468226,
|
| 287 |
+
"loss": 2.2381,
|
| 288 |
+
"step": 22000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.0,
|
| 292 |
+
"learning_rate": 0.0025123745819397992,
|
| 293 |
+
"loss": 2.2507,
|
| 294 |
+
"step": 22500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.0,
|
| 298 |
+
"eval_loss": 2.156663179397583,
|
| 299 |
+
"eval_runtime": 12.0987,
|
| 300 |
+
"eval_samples_per_second": 41.327,
|
| 301 |
+
"eval_steps_per_second": 41.327,
|
| 302 |
+
"step": 22500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.07,
|
| 306 |
+
"learning_rate": 0.002345150501672241,
|
| 307 |
+
"loss": 2.2693,
|
| 308 |
+
"step": 23000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.13,
|
| 312 |
+
"learning_rate": 0.002177926421404682,
|
| 313 |
+
"loss": 2.2605,
|
| 314 |
+
"step": 23500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.2,
|
| 318 |
+
"learning_rate": 0.002010702341137124,
|
| 319 |
+
"loss": 2.2272,
|
| 320 |
+
"step": 24000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 3.27,
|
| 324 |
+
"learning_rate": 0.0018434782608695653,
|
| 325 |
+
"loss": 2.2394,
|
| 326 |
+
"step": 24500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 3.33,
|
| 330 |
+
"learning_rate": 0.0016765886287625417,
|
| 331 |
+
"loss": 2.231,
|
| 332 |
+
"step": 25000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.4,
|
| 336 |
+
"learning_rate": 0.0015093645484949834,
|
| 337 |
+
"loss": 2.225,
|
| 338 |
+
"step": 25500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.47,
|
| 342 |
+
"learning_rate": 0.0013421404682274248,
|
| 343 |
+
"loss": 2.2503,
|
| 344 |
+
"step": 26000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.53,
|
| 348 |
+
"learning_rate": 0.0011749163879598663,
|
| 349 |
+
"loss": 2.2597,
|
| 350 |
+
"step": 26500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.6,
|
| 354 |
+
"learning_rate": 0.001008026755852843,
|
| 355 |
+
"loss": 2.228,
|
| 356 |
+
"step": 27000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 3.67,
|
| 360 |
+
"learning_rate": 0.0008408026755852843,
|
| 361 |
+
"loss": 2.2338,
|
| 362 |
+
"step": 27500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 3.73,
|
| 366 |
+
"learning_rate": 0.0006735785953177258,
|
| 367 |
+
"loss": 2.2478,
|
| 368 |
+
"step": 28000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 3.8,
|
| 372 |
+
"learning_rate": 0.0005063545150501672,
|
| 373 |
+
"loss": 2.2672,
|
| 374 |
+
"step": 28500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 3.87,
|
| 378 |
+
"learning_rate": 0.0003391304347826087,
|
| 379 |
+
"loss": 2.2513,
|
| 380 |
+
"step": 29000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 3.93,
|
| 384 |
+
"learning_rate": 0.0001722408026755853,
|
| 385 |
+
"loss": 2.2776,
|
| 386 |
+
"step": 29500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 4.0,
|
| 390 |
+
"learning_rate": 5.016722408026756e-06,
|
| 391 |
+
"loss": 2.2519,
|
| 392 |
+
"step": 30000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 4.0,
|
| 396 |
+
"eval_loss": 2.1534345149993896,
|
| 397 |
+
"eval_runtime": 11.8174,
|
| 398 |
+
"eval_samples_per_second": 42.31,
|
| 399 |
+
"eval_steps_per_second": 42.31,
|
| 400 |
+
"step": 30000
|
| 401 |
+
}
|
| 402 |
+
],
|
| 403 |
+
"max_steps": 30000,
|
| 404 |
+
"num_train_epochs": 4,
|
| 405 |
+
"total_flos": 3.135504384e+16,
|
| 406 |
+
"trial_name": null,
|
| 407 |
+
"trial_params": null
|
| 408 |
+
}
|
checkpoint-30000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b266296d18d0d931e280725174f6a350cf988e8a9bc67cb4310bffb7b37b68e4
|
| 3 |
+
size 3963
|
checkpoint-30000/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
logs/events.out.tfevents.1690017287.7cfadc3aec44.25659.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a3a9e59b13a6b3e9e4bc2ea3918492050f081552d6c569989168da8189b5ba2
|
| 3 |
+
size 4727
|
logs/events.out.tfevents.1690017422.7cfadc3aec44.25659.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:113c169aa12eb4fa126b746d590aeac94dc5743af1505f8f0d5911b3959f2cc1
|
| 3 |
+
size 4184
|
logs/events.out.tfevents.1690017606.7cfadc3aec44.27769.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e0d4794b5ab90a331be27d19461e4b1560539b3ba862b869aee9819c3925cbf
|
| 3 |
+
size 4184
|
logs/events.out.tfevents.1690017757.7cfadc3aec44.28627.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4de0a6adb97f4a45795204ba92f2bd793f74d6ba27a1d3abd6956adfa62ea259
|
| 3 |
+
size 15371
|
logs/events.out.tfevents.1690021418.7cfadc3aec44.28627.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a9e8356cbaf333541a81adae204b5decc69dd8026cf1662f9dafbf5133d8e22
|
| 3 |
+
size 364
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"### End",
|
| 4 |
+
"### Instruction:",
|
| 5 |
+
"### Response:\n"
|
| 6 |
+
],
|
| 7 |
+
"bos_token": "<|endoftext|>",
|
| 8 |
+
"eos_token": "<|endoftext|>",
|
| 9 |
+
"pad_token": "<|endoftext|>",
|
| 10 |
+
"unk_token": "<|endoftext|>"
|
| 11 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"bos_token": {
|
| 5 |
+
"__type": "AddedToken",
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"clean_up_tokenization_spaces": true,
|
| 13 |
+
"eos_token": {
|
| 14 |
+
"__type": "AddedToken",
|
| 15 |
+
"content": "<|endoftext|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": true,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false
|
| 20 |
+
},
|
| 21 |
+
"errors": "replace",
|
| 22 |
+
"max_seq_length": 512,
|
| 23 |
+
"model_max_length": 1024,
|
| 24 |
+
"pad_token": null,
|
| 25 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 26 |
+
"unk_token": {
|
| 27 |
+
"__type": "AddedToken",
|
| 28 |
+
"content": "<|endoftext|>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": true,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false
|
| 33 |
+
},
|
| 34 |
+
"use_fast": false
|
| 35 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|