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 +379 -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-warm-start
|
| 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-warm-start
|
| 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-warm-start", 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/uuc18xow)
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.0
|
| 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": 10394912686080.0,
|
| 4 |
+
"train_loss": 0.3196908062759866,
|
| 5 |
+
"train_runtime": 717.6944,
|
| 6 |
+
"train_samples": 1485,
|
| 7 |
+
"train_samples_per_second": 2.069,
|
| 8 |
+
"train_steps_per_second": 0.065
|
| 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": 10394912686080.0,
|
| 4 |
+
"train_loss": 0.3196908062759866,
|
| 5 |
+
"train_runtime": 717.6944,
|
| 6 |
+
"train_samples": 1485,
|
| 7 |
+
"train_samples_per_second": 2.069,
|
| 8 |
+
"train_steps_per_second": 0.065
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,379 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 47,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02127659574468085,
|
| 13 |
+
"grad_norm": 173.3797693640036,
|
| 14 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
+
"loss": 2.0421,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0425531914893617,
|
| 20 |
+
"grad_norm": 213.4863387447492,
|
| 21 |
+
"learning_rate": 4.000000000000001e-06,
|
| 22 |
+
"loss": 2.1926,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06382978723404255,
|
| 27 |
+
"grad_norm": 139.12102571177553,
|
| 28 |
+
"learning_rate": 6e-06,
|
| 29 |
+
"loss": 1.6796,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0851063829787234,
|
| 34 |
+
"grad_norm": 31.770887201034693,
|
| 35 |
+
"learning_rate": 8.000000000000001e-06,
|
| 36 |
+
"loss": 0.8817,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.10638297872340426,
|
| 41 |
+
"grad_norm": 11.974572830777873,
|
| 42 |
+
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 0.6561,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1276595744680851,
|
| 48 |
+
"grad_norm": 5.587929077588245,
|
| 49 |
+
"learning_rate": 9.986018985905901e-06,
|
| 50 |
+
"loss": 0.4698,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14893617021276595,
|
| 55 |
+
"grad_norm": 7.58893303595288,
|
| 56 |
+
"learning_rate": 9.944154131125643e-06,
|
| 57 |
+
"loss": 0.3525,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1702127659574468,
|
| 62 |
+
"grad_norm": 4.03951296758354,
|
| 63 |
+
"learning_rate": 9.874639560909118e-06,
|
| 64 |
+
"loss": 0.2916,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.19148936170212766,
|
| 69 |
+
"grad_norm": 2.36253683449773,
|
| 70 |
+
"learning_rate": 9.777864028930705e-06,
|
| 71 |
+
"loss": 0.2413,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2127659574468085,
|
| 76 |
+
"grad_norm": 1.724376767595317,
|
| 77 |
+
"learning_rate": 9.654368743221022e-06,
|
| 78 |
+
"loss": 0.2407,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.23404255319148937,
|
| 83 |
+
"grad_norm": 7.656722082252226,
|
| 84 |
+
"learning_rate": 9.504844339512096e-06,
|
| 85 |
+
"loss": 0.2365,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2553191489361702,
|
| 90 |
+
"grad_norm": 3.966824389729543,
|
| 91 |
+
"learning_rate": 9.330127018922195e-06,
|
| 92 |
+
"loss": 0.2593,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2765957446808511,
|
| 97 |
+
"grad_norm": 2.4259726258566894,
|
| 98 |
+
"learning_rate": 9.131193871579975e-06,
|
| 99 |
+
"loss": 0.2063,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.2978723404255319,
|
| 104 |
+
"grad_norm": 2.0989965880083665,
|
| 105 |
+
"learning_rate": 8.90915741234015e-06,
|
| 106 |
+
"loss": 0.2137,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3191489361702128,
|
| 111 |
+
"grad_norm": 0.7473125600483845,
|
| 112 |
+
"learning_rate": 8.665259359149132e-06,
|
| 113 |
+
"loss": 0.1752,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3404255319148936,
|
| 118 |
+
"grad_norm": 2.362035437648623,
|
| 119 |
+
"learning_rate": 8.400863688854598e-06,
|
| 120 |
+
"loss": 0.2079,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3617021276595745,
|
| 125 |
+
"grad_norm": 1.1278643764919287,
|
| 126 |
+
"learning_rate": 8.117449009293668e-06,
|
| 127 |
+
"loss": 0.2096,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3829787234042553,
|
| 132 |
+
"grad_norm": 0.9888933504092371,
|
| 133 |
+
"learning_rate": 7.81660029031811e-06,
|
| 134 |
+
"loss": 0.1759,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.40425531914893614,
|
| 139 |
+
"grad_norm": 1.3935821371981716,
|
| 140 |
+
"learning_rate": 7.500000000000001e-06,
|
| 141 |
+
"loss": 0.2036,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.425531914893617,
|
| 146 |
+
"grad_norm": 0.7210016827193401,
|
| 147 |
+
"learning_rate": 7.169418695587791e-06,
|
| 148 |
+
"loss": 0.1757,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.44680851063829785,
|
| 153 |
+
"grad_norm": 1.3893738922334362,
|
| 154 |
+
"learning_rate": 6.8267051218319766e-06,
|
| 155 |
+
"loss": 0.1802,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.46808510638297873,
|
| 160 |
+
"grad_norm": 0.7826903349267914,
|
| 161 |
+
"learning_rate": 6.473775872054522e-06,
|
| 162 |
+
"loss": 0.1823,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.48936170212765956,
|
| 167 |
+
"grad_norm": 1.1055685574886962,
|
| 168 |
+
"learning_rate": 6.112604669781572e-06,
|
| 169 |
+
"loss": 0.1812,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5106382978723404,
|
| 174 |
+
"grad_norm": 1.3398319352785983,
|
| 175 |
+
"learning_rate": 5.745211330880872e-06,
|
| 176 |
+
"loss": 0.1791,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5319148936170213,
|
| 181 |
+
"grad_norm": 0.5530788932061305,
|
| 182 |
+
"learning_rate": 5.373650467932122e-06,
|
| 183 |
+
"loss": 0.1582,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5531914893617021,
|
| 188 |
+
"grad_norm": 0.979470403637453,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.1687,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.574468085106383,
|
| 195 |
+
"grad_norm": 0.6524782994513246,
|
| 196 |
+
"learning_rate": 4.626349532067879e-06,
|
| 197 |
+
"loss": 0.1614,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5957446808510638,
|
| 202 |
+
"grad_norm": 0.613082643775453,
|
| 203 |
+
"learning_rate": 4.254788669119127e-06,
|
| 204 |
+
"loss": 0.1406,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6170212765957447,
|
| 209 |
+
"grad_norm": 0.6117099956858949,
|
| 210 |
+
"learning_rate": 3.887395330218429e-06,
|
| 211 |
+
"loss": 0.1637,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6382978723404256,
|
| 216 |
+
"grad_norm": 0.4598315067554906,
|
| 217 |
+
"learning_rate": 3.526224127945479e-06,
|
| 218 |
+
"loss": 0.1488,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6595744680851063,
|
| 223 |
+
"grad_norm": 0.5041452368752941,
|
| 224 |
+
"learning_rate": 3.173294878168025e-06,
|
| 225 |
+
"loss": 0.1385,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6808510638297872,
|
| 230 |
+
"grad_norm": 0.6131569474074337,
|
| 231 |
+
"learning_rate": 2.83058130441221e-06,
|
| 232 |
+
"loss": 0.151,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7021276595744681,
|
| 237 |
+
"grad_norm": 0.5119335095129718,
|
| 238 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 239 |
+
"loss": 0.1492,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.723404255319149,
|
| 244 |
+
"grad_norm": 0.4443782774575828,
|
| 245 |
+
"learning_rate": 2.1833997096818897e-06,
|
| 246 |
+
"loss": 0.1174,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7446808510638298,
|
| 251 |
+
"grad_norm": 0.45664829579358884,
|
| 252 |
+
"learning_rate": 1.8825509907063328e-06,
|
| 253 |
+
"loss": 0.1273,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7659574468085106,
|
| 258 |
+
"grad_norm": 0.5213080660032051,
|
| 259 |
+
"learning_rate": 1.5991363111454023e-06,
|
| 260 |
+
"loss": 0.1422,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7872340425531915,
|
| 265 |
+
"grad_norm": 0.5400795029678803,
|
| 266 |
+
"learning_rate": 1.3347406408508695e-06,
|
| 267 |
+
"loss": 0.148,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.8085106382978723,
|
| 272 |
+
"grad_norm": 0.47711498862269414,
|
| 273 |
+
"learning_rate": 1.0908425876598512e-06,
|
| 274 |
+
"loss": 0.1201,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8297872340425532,
|
| 279 |
+
"grad_norm": 0.4674822262301143,
|
| 280 |
+
"learning_rate": 8.688061284200266e-07,
|
| 281 |
+
"loss": 0.1427,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.851063829787234,
|
| 286 |
+
"grad_norm": 0.4740692779208992,
|
| 287 |
+
"learning_rate": 6.698729810778065e-07,
|
| 288 |
+
"loss": 0.1259,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.8723404255319149,
|
| 293 |
+
"grad_norm": 0.5372156644144523,
|
| 294 |
+
"learning_rate": 4.951556604879049e-07,
|
| 295 |
+
"loss": 0.1342,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8936170212765957,
|
| 300 |
+
"grad_norm": 0.5807354531086322,
|
| 301 |
+
"learning_rate": 3.4563125677897936e-07,
|
| 302 |
+
"loss": 0.1449,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.9148936170212766,
|
| 307 |
+
"grad_norm": 0.4917053466958668,
|
| 308 |
+
"learning_rate": 2.2213597106929608e-07,
|
| 309 |
+
"loss": 0.1344,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9361702127659575,
|
| 314 |
+
"grad_norm": 0.45294061142670444,
|
| 315 |
+
"learning_rate": 1.253604390908819e-07,
|
| 316 |
+
"loss": 0.1187,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9574468085106383,
|
| 321 |
+
"grad_norm": 0.42691924060512204,
|
| 322 |
+
"learning_rate": 5.584586887435739e-08,
|
| 323 |
+
"loss": 0.1114,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.9787234042553191,
|
| 328 |
+
"grad_norm": 0.4007354051634515,
|
| 329 |
+
"learning_rate": 1.3981014094099354e-08,
|
| 330 |
+
"loss": 0.1139,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.0,
|
| 335 |
+
"grad_norm": 0.4672720715521086,
|
| 336 |
+
"learning_rate": 0.0,
|
| 337 |
+
"loss": 0.1298,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.0,
|
| 342 |
+
"eval_loss": 0.14448508620262146,
|
| 343 |
+
"eval_runtime": 6.6249,
|
| 344 |
+
"eval_samples_per_second": 11.925,
|
| 345 |
+
"eval_steps_per_second": 3.019,
|
| 346 |
+
"step": 47
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.0,
|
| 350 |
+
"step": 47,
|
| 351 |
+
"total_flos": 10394912686080.0,
|
| 352 |
+
"train_loss": 0.3196908062759866,
|
| 353 |
+
"train_runtime": 717.6944,
|
| 354 |
+
"train_samples_per_second": 2.069,
|
| 355 |
+
"train_steps_per_second": 0.065
|
| 356 |
+
}
|
| 357 |
+
],
|
| 358 |
+
"logging_steps": 1,
|
| 359 |
+
"max_steps": 47,
|
| 360 |
+
"num_input_tokens_seen": 0,
|
| 361 |
+
"num_train_epochs": 1,
|
| 362 |
+
"save_steps": 500,
|
| 363 |
+
"stateful_callbacks": {
|
| 364 |
+
"TrainerControl": {
|
| 365 |
+
"args": {
|
| 366 |
+
"should_epoch_stop": false,
|
| 367 |
+
"should_evaluate": false,
|
| 368 |
+
"should_log": false,
|
| 369 |
+
"should_save": true,
|
| 370 |
+
"should_training_stop": true
|
| 371 |
+
},
|
| 372 |
+
"attributes": {}
|
| 373 |
+
}
|
| 374 |
+
},
|
| 375 |
+
"total_flos": 10394912686080.0,
|
| 376 |
+
"train_batch_size": 8,
|
| 377 |
+
"trial_name": null,
|
| 378 |
+
"trial_params": null
|
| 379 |
+
}
|