Commit
•
bd1255c
1
Parent(s):
cb17dde
Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- generation_config.json +12 -0
- train_results.json +9 -0
- trainer_state.json +400 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: barc0/Llama-3.1-ARC-Potpourri-Induction-8B
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: potpourri-8b-inst-fft-induction-bc-trajectory-max10-per-task-check-answer
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for potpourri-8b-inst-fft-induction-bc-trajectory-max10-per-task-check-answer
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [barc0/Llama-3.1-ARC-Potpourri-Induction-8B](https://huggingface.co/barc0/Llama-3.1-ARC-Potpourri-Induction-8B).
|
| 15 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
+
|
| 17 |
+
## Quick start
|
| 18 |
+
|
| 19 |
+
```python
|
| 20 |
+
from transformers import pipeline
|
| 21 |
+
|
| 22 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 23 |
+
generator = pipeline("text-generation", model="suehyunpark/potpourri-8b-inst-fft-induction-bc-trajectory-max10-per-task-check-answer", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/suehyun/arc-improve/runs/4y71yzaa)
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.1
|
| 37 |
+
- Transformers: 4.46.2
|
| 38 |
+
- Pytorch: 2.5.1
|
| 39 |
+
- Datasets: 3.1.0
|
| 40 |
+
- Tokenizers: 0.20.3
|
| 41 |
+
|
| 42 |
+
## Citations
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
Cite TRL as:
|
| 47 |
+
|
| 48 |
+
```bibtex
|
| 49 |
+
@misc{vonwerra2022trl,
|
| 50 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 51 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
|
| 52 |
+
year = 2020,
|
| 53 |
+
journal = {GitHub repository},
|
| 54 |
+
publisher = {GitHub},
|
| 55 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 56 |
+
}
|
| 57 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 11109249515520.0,
|
| 4 |
+
"train_loss": 0.31718048006296157,
|
| 5 |
+
"train_runtime": 911.8037,
|
| 6 |
+
"train_samples": 1580,
|
| 7 |
+
"train_samples_per_second": 1.733,
|
| 8 |
+
"train_steps_per_second": 0.055
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 128000,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
128001,
|
| 6 |
+
128008,
|
| 7 |
+
128009
|
| 8 |
+
],
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "4.46.2"
|
| 12 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 11109249515520.0,
|
| 4 |
+
"train_loss": 0.31718048006296157,
|
| 5 |
+
"train_runtime": 911.8037,
|
| 6 |
+
"train_samples": 1580,
|
| 7 |
+
"train_samples_per_second": 1.733,
|
| 8 |
+
"train_steps_per_second": 0.055
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,400 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 50,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 177.15509899472193,
|
| 14 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
+
"loss": 2.0908,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04,
|
| 20 |
+
"grad_norm": 249.41431883318035,
|
| 21 |
+
"learning_rate": 4.000000000000001e-06,
|
| 22 |
+
"loss": 2.5319,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06,
|
| 27 |
+
"grad_norm": 150.39879752385414,
|
| 28 |
+
"learning_rate": 6e-06,
|
| 29 |
+
"loss": 1.7308,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08,
|
| 34 |
+
"grad_norm": 37.728760587541345,
|
| 35 |
+
"learning_rate": 8.000000000000001e-06,
|
| 36 |
+
"loss": 0.9312,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1,
|
| 41 |
+
"grad_norm": 15.70356057623236,
|
| 42 |
+
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 0.7705,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12,
|
| 48 |
+
"grad_norm": 5.45224867064269,
|
| 49 |
+
"learning_rate": 9.987820251299121e-06,
|
| 50 |
+
"loss": 0.4923,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14,
|
| 55 |
+
"grad_norm": 4.908282773077112,
|
| 56 |
+
"learning_rate": 9.951340343707852e-06,
|
| 57 |
+
"loss": 0.3728,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.16,
|
| 62 |
+
"grad_norm": 3.502796679759215,
|
| 63 |
+
"learning_rate": 9.890738003669029e-06,
|
| 64 |
+
"loss": 0.2617,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.18,
|
| 69 |
+
"grad_norm": 1.7024726367446046,
|
| 70 |
+
"learning_rate": 9.806308479691595e-06,
|
| 71 |
+
"loss": 0.2508,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2,
|
| 76 |
+
"grad_norm": 0.9940466279699945,
|
| 77 |
+
"learning_rate": 9.698463103929542e-06,
|
| 78 |
+
"loss": 0.2137,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.22,
|
| 83 |
+
"grad_norm": 2.450501035506251,
|
| 84 |
+
"learning_rate": 9.567727288213005e-06,
|
| 85 |
+
"loss": 0.2401,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.24,
|
| 90 |
+
"grad_norm": 1.4427281883701062,
|
| 91 |
+
"learning_rate": 9.414737964294636e-06,
|
| 92 |
+
"loss": 0.2037,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.26,
|
| 97 |
+
"grad_norm": 0.904196965125475,
|
| 98 |
+
"learning_rate": 9.24024048078213e-06,
|
| 99 |
+
"loss": 0.1882,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.28,
|
| 104 |
+
"grad_norm": 0.9298794223496288,
|
| 105 |
+
"learning_rate": 9.045084971874738e-06,
|
| 106 |
+
"loss": 0.1844,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3,
|
| 111 |
+
"grad_norm": 1.855575170826518,
|
| 112 |
+
"learning_rate": 8.83022221559489e-06,
|
| 113 |
+
"loss": 0.1994,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.32,
|
| 118 |
+
"grad_norm": 0.6710757494156826,
|
| 119 |
+
"learning_rate": 8.596699001693257e-06,
|
| 120 |
+
"loss": 0.1774,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.34,
|
| 125 |
+
"grad_norm": 2.329770545062615,
|
| 126 |
+
"learning_rate": 8.345653031794292e-06,
|
| 127 |
+
"loss": 0.1916,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.36,
|
| 132 |
+
"grad_norm": 0.928554494466913,
|
| 133 |
+
"learning_rate": 8.078307376628292e-06,
|
| 134 |
+
"loss": 0.1859,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.38,
|
| 139 |
+
"grad_norm": 0.9443531833751477,
|
| 140 |
+
"learning_rate": 7.795964517353734e-06,
|
| 141 |
+
"loss": 0.1699,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.4,
|
| 146 |
+
"grad_norm": 0.7918115530539849,
|
| 147 |
+
"learning_rate": 7.500000000000001e-06,
|
| 148 |
+
"loss": 0.1783,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.42,
|
| 153 |
+
"grad_norm": 0.9196163123591217,
|
| 154 |
+
"learning_rate": 7.191855733945388e-06,
|
| 155 |
+
"loss": 0.167,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.44,
|
| 160 |
+
"grad_norm": 0.7595897977250279,
|
| 161 |
+
"learning_rate": 6.873032967079562e-06,
|
| 162 |
+
"loss": 0.1778,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.46,
|
| 167 |
+
"grad_norm": 0.6745291224887299,
|
| 168 |
+
"learning_rate": 6.545084971874738e-06,
|
| 169 |
+
"loss": 0.1546,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.48,
|
| 174 |
+
"grad_norm": 0.8011676613761106,
|
| 175 |
+
"learning_rate": 6.209609477998339e-06,
|
| 176 |
+
"loss": 0.1674,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5,
|
| 181 |
+
"grad_norm": 0.5793616119081132,
|
| 182 |
+
"learning_rate": 5.8682408883346535e-06,
|
| 183 |
+
"loss": 0.1489,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.52,
|
| 188 |
+
"grad_norm": 0.502456080126672,
|
| 189 |
+
"learning_rate": 5.522642316338268e-06,
|
| 190 |
+
"loss": 0.1673,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.54,
|
| 195 |
+
"grad_norm": 0.9029577834941641,
|
| 196 |
+
"learning_rate": 5.174497483512506e-06,
|
| 197 |
+
"loss": 0.1562,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.56,
|
| 202 |
+
"grad_norm": 0.5880447415239002,
|
| 203 |
+
"learning_rate": 4.825502516487497e-06,
|
| 204 |
+
"loss": 0.1538,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.58,
|
| 209 |
+
"grad_norm": 0.6959640251857308,
|
| 210 |
+
"learning_rate": 4.477357683661734e-06,
|
| 211 |
+
"loss": 0.1317,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6,
|
| 216 |
+
"grad_norm": 0.819167997187879,
|
| 217 |
+
"learning_rate": 4.131759111665349e-06,
|
| 218 |
+
"loss": 0.159,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.62,
|
| 223 |
+
"grad_norm": 0.700213501190347,
|
| 224 |
+
"learning_rate": 3.790390522001662e-06,
|
| 225 |
+
"loss": 0.1332,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.64,
|
| 230 |
+
"grad_norm": 0.5453470463406256,
|
| 231 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 232 |
+
"loss": 0.1493,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.66,
|
| 237 |
+
"grad_norm": 0.7267182413675126,
|
| 238 |
+
"learning_rate": 3.12696703292044e-06,
|
| 239 |
+
"loss": 0.1449,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.68,
|
| 244 |
+
"grad_norm": 0.6201985145724069,
|
| 245 |
+
"learning_rate": 2.8081442660546126e-06,
|
| 246 |
+
"loss": 0.1489,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7,
|
| 251 |
+
"grad_norm": 0.6063378038267555,
|
| 252 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 253 |
+
"loss": 0.1374,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.72,
|
| 258 |
+
"grad_norm": 0.6864441408653645,
|
| 259 |
+
"learning_rate": 2.204035482646267e-06,
|
| 260 |
+
"loss": 0.1326,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.74,
|
| 265 |
+
"grad_norm": 0.5642913934479935,
|
| 266 |
+
"learning_rate": 1.9216926233717087e-06,
|
| 267 |
+
"loss": 0.1117,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.76,
|
| 272 |
+
"grad_norm": 0.5716062286734344,
|
| 273 |
+
"learning_rate": 1.6543469682057105e-06,
|
| 274 |
+
"loss": 0.1294,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.78,
|
| 279 |
+
"grad_norm": 0.7133036398240772,
|
| 280 |
+
"learning_rate": 1.4033009983067454e-06,
|
| 281 |
+
"loss": 0.1449,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.8,
|
| 286 |
+
"grad_norm": 0.5522465024093508,
|
| 287 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 288 |
+
"loss": 0.1286,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.82,
|
| 293 |
+
"grad_norm": 0.5011972438671921,
|
| 294 |
+
"learning_rate": 9.549150281252633e-07,
|
| 295 |
+
"loss": 0.1347,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.84,
|
| 300 |
+
"grad_norm": 0.6161564938944896,
|
| 301 |
+
"learning_rate": 7.597595192178702e-07,
|
| 302 |
+
"loss": 0.1453,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.86,
|
| 307 |
+
"grad_norm": 0.6818018896853675,
|
| 308 |
+
"learning_rate": 5.852620357053651e-07,
|
| 309 |
+
"loss": 0.147,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.88,
|
| 314 |
+
"grad_norm": 0.6378587677381994,
|
| 315 |
+
"learning_rate": 4.322727117869951e-07,
|
| 316 |
+
"loss": 0.138,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9,
|
| 321 |
+
"grad_norm": 0.47712509769630884,
|
| 322 |
+
"learning_rate": 3.015368960704584e-07,
|
| 323 |
+
"loss": 0.1273,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.92,
|
| 328 |
+
"grad_norm": 0.5347492598335275,
|
| 329 |
+
"learning_rate": 1.9369152030840553e-07,
|
| 330 |
+
"loss": 0.1372,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.94,
|
| 335 |
+
"grad_norm": 0.4147667548401041,
|
| 336 |
+
"learning_rate": 1.0926199633097156e-07,
|
| 337 |
+
"loss": 0.1365,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.96,
|
| 342 |
+
"grad_norm": 0.45562021103374617,
|
| 343 |
+
"learning_rate": 4.865965629214819e-08,
|
| 344 |
+
"loss": 0.1365,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.98,
|
| 349 |
+
"grad_norm": 0.4580789267861981,
|
| 350 |
+
"learning_rate": 1.2179748700879013e-08,
|
| 351 |
+
"loss": 0.1165,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.0,
|
| 356 |
+
"grad_norm": 0.4569340519935219,
|
| 357 |
+
"learning_rate": 0.0,
|
| 358 |
+
"loss": 0.1302,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.0,
|
| 363 |
+
"eval_loss": 0.1237141489982605,
|
| 364 |
+
"eval_runtime": 7.208,
|
| 365 |
+
"eval_samples_per_second": 11.654,
|
| 366 |
+
"eval_steps_per_second": 2.913,
|
| 367 |
+
"step": 50
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.0,
|
| 371 |
+
"step": 50,
|
| 372 |
+
"total_flos": 11109249515520.0,
|
| 373 |
+
"train_loss": 0.31718048006296157,
|
| 374 |
+
"train_runtime": 911.8037,
|
| 375 |
+
"train_samples_per_second": 1.733,
|
| 376 |
+
"train_steps_per_second": 0.055
|
| 377 |
+
}
|
| 378 |
+
],
|
| 379 |
+
"logging_steps": 1,
|
| 380 |
+
"max_steps": 50,
|
| 381 |
+
"num_input_tokens_seen": 0,
|
| 382 |
+
"num_train_epochs": 1,
|
| 383 |
+
"save_steps": 500,
|
| 384 |
+
"stateful_callbacks": {
|
| 385 |
+
"TrainerControl": {
|
| 386 |
+
"args": {
|
| 387 |
+
"should_epoch_stop": false,
|
| 388 |
+
"should_evaluate": false,
|
| 389 |
+
"should_log": false,
|
| 390 |
+
"should_save": true,
|
| 391 |
+
"should_training_stop": true
|
| 392 |
+
},
|
| 393 |
+
"attributes": {}
|
| 394 |
+
}
|
| 395 |
+
},
|
| 396 |
+
"total_flos": 11109249515520.0,
|
| 397 |
+
"train_batch_size": 8,
|
| 398 |
+
"trial_name": null,
|
| 399 |
+
"trial_params": null
|
| 400 |
+
}
|