Model save
Browse files- README.md +58 -0
- all_results.json +8 -0
- generation_config.json +9 -0
- train_results.json +8 -0
- trainer_state.json +539 -0
README.md
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen2.5-Math-1.5B-Instruct
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Qwen2.5-1.5B-Math-Instruct-LIMO
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Qwen2.5-1.5B-Math-Instruct-LIMO
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-Math-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B-Instruct).
|
| 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="haoyuw/Qwen2.5-1.5B-Math-Instruct-LIMO", 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 |
+
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with SFT.
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.16.0.dev0
|
| 38 |
+
- Transformers: 4.49.0.dev0
|
| 39 |
+
- Pytorch: 2.5.1
|
| 40 |
+
- Datasets: 3.2.0
|
| 41 |
+
- Tokenizers: 0.21.0
|
| 42 |
+
|
| 43 |
+
## Citations
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
Cite TRL as:
|
| 48 |
+
|
| 49 |
+
```bibtex
|
| 50 |
+
@misc{vonwerra2022trl,
|
| 51 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 52 |
+
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},
|
| 53 |
+
year = 2020,
|
| 54 |
+
journal = {GitHub repository},
|
| 55 |
+
publisher = {GitHub},
|
| 56 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 57 |
+
}
|
| 58 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 20283251490816.0,
|
| 3 |
+
"train_loss": 0.7583397939968645,
|
| 4 |
+
"train_runtime": 1935.5651,
|
| 5 |
+
"train_samples": 817,
|
| 6 |
+
"train_samples_per_second": 11.753,
|
| 7 |
+
"train_steps_per_second": 0.184
|
| 8 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"transformers_version": "4.49.0.dev0"
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_flos": 20283251490816.0,
|
| 3 |
+
"train_loss": 0.7583397939968645,
|
| 4 |
+
"train_runtime": 1935.5651,
|
| 5 |
+
"train_samples": 817,
|
| 6 |
+
"train_samples_per_second": 11.753,
|
| 7 |
+
"train_steps_per_second": 0.184
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,539 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 356,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.056179775280898875,
|
| 13 |
+
"grad_norm": 1.6944109649364811,
|
| 14 |
+
"learning_rate": 5.555555555555557e-06,
|
| 15 |
+
"loss": 1.3057,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11235955056179775,
|
| 20 |
+
"grad_norm": 1.0734576025480336,
|
| 21 |
+
"learning_rate": 1.1111111111111113e-05,
|
| 22 |
+
"loss": 1.2523,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.16853932584269662,
|
| 27 |
+
"grad_norm": 0.9680158843960001,
|
| 28 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 29 |
+
"loss": 1.1501,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2247191011235955,
|
| 34 |
+
"grad_norm": 0.8622223090177016,
|
| 35 |
+
"learning_rate": 1.9998445014500246e-05,
|
| 36 |
+
"loss": 1.0298,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2808988764044944,
|
| 41 |
+
"grad_norm": 0.7503899959880194,
|
| 42 |
+
"learning_rate": 1.998095759792783e-05,
|
| 43 |
+
"loss": 1.0131,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.33707865168539325,
|
| 48 |
+
"grad_norm": 0.5972359554952426,
|
| 49 |
+
"learning_rate": 1.9944076920747774e-05,
|
| 50 |
+
"loss": 0.9485,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.39325842696629215,
|
| 55 |
+
"grad_norm": 0.5365600358115739,
|
| 56 |
+
"learning_rate": 1.98878826221941e-05,
|
| 57 |
+
"loss": 0.9524,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.449438202247191,
|
| 62 |
+
"grad_norm": 0.4710678595191818,
|
| 63 |
+
"learning_rate": 1.9812496046863375e-05,
|
| 64 |
+
"loss": 0.9382,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.5056179775280899,
|
| 69 |
+
"grad_norm": 0.44555673749010527,
|
| 70 |
+
"learning_rate": 1.971807998268615e-05,
|
| 71 |
+
"loss": 0.91,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5617977528089888,
|
| 76 |
+
"grad_norm": 0.4474821677236859,
|
| 77 |
+
"learning_rate": 1.9604838309406668e-05,
|
| 78 |
+
"loss": 0.8664,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.6179775280898876,
|
| 83 |
+
"grad_norm": 0.46993823824767494,
|
| 84 |
+
"learning_rate": 1.947301555832993e-05,
|
| 85 |
+
"loss": 0.8943,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.6741573033707865,
|
| 90 |
+
"grad_norm": 0.40887692660248126,
|
| 91 |
+
"learning_rate": 1.932289638428676e-05,
|
| 92 |
+
"loss": 0.8427,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.7303370786516854,
|
| 97 |
+
"grad_norm": 0.4178479209299752,
|
| 98 |
+
"learning_rate": 1.9154804950957096e-05,
|
| 99 |
+
"loss": 0.8765,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.7865168539325843,
|
| 104 |
+
"grad_norm": 0.4358655706142232,
|
| 105 |
+
"learning_rate": 1.896910423087889e-05,
|
| 106 |
+
"loss": 0.8815,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.8426966292134831,
|
| 111 |
+
"grad_norm": 0.4759614764741585,
|
| 112 |
+
"learning_rate": 1.8766195221654038e-05,
|
| 113 |
+
"loss": 0.9062,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.898876404494382,
|
| 118 |
+
"grad_norm": 0.40941475449933445,
|
| 119 |
+
"learning_rate": 1.8546516080043956e-05,
|
| 120 |
+
"loss": 0.8525,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.9550561797752809,
|
| 125 |
+
"grad_norm": 0.46163800722604575,
|
| 126 |
+
"learning_rate": 1.831054117582457e-05,
|
| 127 |
+
"loss": 0.8511,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.0112359550561798,
|
| 132 |
+
"grad_norm": 0.4144691948002747,
|
| 133 |
+
"learning_rate": 1.805878006744379e-05,
|
| 134 |
+
"loss": 0.8351,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.0674157303370786,
|
| 139 |
+
"grad_norm": 0.41318674441336783,
|
| 140 |
+
"learning_rate": 1.7791776401693455e-05,
|
| 141 |
+
"loss": 0.7575,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.1235955056179776,
|
| 146 |
+
"grad_norm": 0.4140804142512328,
|
| 147 |
+
"learning_rate": 1.7510106739771698e-05,
|
| 148 |
+
"loss": 0.7629,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.1797752808988764,
|
| 153 |
+
"grad_norm": 0.3888169360375874,
|
| 154 |
+
"learning_rate": 1.7214379312270786e-05,
|
| 155 |
+
"loss": 0.7725,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.2359550561797752,
|
| 160 |
+
"grad_norm": 0.39830672831462005,
|
| 161 |
+
"learning_rate": 1.690523270577883e-05,
|
| 162 |
+
"loss": 0.7957,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.2921348314606742,
|
| 167 |
+
"grad_norm": 0.4025052364368211,
|
| 168 |
+
"learning_rate": 1.6583334483931465e-05,
|
| 169 |
+
"loss": 0.8151,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.348314606741573,
|
| 174 |
+
"grad_norm": 0.40954244286961317,
|
| 175 |
+
"learning_rate": 1.6249379745891277e-05,
|
| 176 |
+
"loss": 0.7625,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.404494382022472,
|
| 181 |
+
"grad_norm": 0.38363991825348154,
|
| 182 |
+
"learning_rate": 1.590408962536759e-05,
|
| 183 |
+
"loss": 0.768,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.4606741573033708,
|
| 188 |
+
"grad_norm": 0.4202167638100542,
|
| 189 |
+
"learning_rate": 1.5548209733417974e-05,
|
| 190 |
+
"loss": 0.7333,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 1.5168539325842696,
|
| 195 |
+
"grad_norm": 0.39549093654722417,
|
| 196 |
+
"learning_rate": 1.5182508548393917e-05,
|
| 197 |
+
"loss": 0.7561,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.5730337078651684,
|
| 202 |
+
"grad_norm": 0.38558554907475995,
|
| 203 |
+
"learning_rate": 1.4807775756507423e-05,
|
| 204 |
+
"loss": 0.7659,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.6292134831460674,
|
| 209 |
+
"grad_norm": 0.40811098318454503,
|
| 210 |
+
"learning_rate": 1.4424820546601892e-05,
|
| 211 |
+
"loss": 0.7592,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.6853932584269664,
|
| 216 |
+
"grad_norm": 0.4075116805060462,
|
| 217 |
+
"learning_rate": 1.4034469862809474e-05,
|
| 218 |
+
"loss": 0.7738,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.7415730337078652,
|
| 223 |
+
"grad_norm": 0.4178748191013212,
|
| 224 |
+
"learning_rate": 1.363756661886812e-05,
|
| 225 |
+
"loss": 0.7677,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.797752808988764,
|
| 230 |
+
"grad_norm": 0.41587042744707226,
|
| 231 |
+
"learning_rate": 1.32349678779542e-05,
|
| 232 |
+
"loss": 0.7555,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.8539325842696628,
|
| 237 |
+
"grad_norm": 0.4186538870477955,
|
| 238 |
+
"learning_rate": 1.282754300196122e-05,
|
| 239 |
+
"loss": 0.7731,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.9101123595505618,
|
| 244 |
+
"grad_norm": 0.40473262826860434,
|
| 245 |
+
"learning_rate": 1.2416171774220976e-05,
|
| 246 |
+
"loss": 0.7644,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.9662921348314608,
|
| 251 |
+
"grad_norm": 0.38178573402418625,
|
| 252 |
+
"learning_rate": 1.2001742499720888e-05,
|
| 253 |
+
"loss": 0.7687,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.0224719101123596,
|
| 258 |
+
"grad_norm": 0.3819894080787757,
|
| 259 |
+
"learning_rate": 1.1585150086919896e-05,
|
| 260 |
+
"loss": 0.7067,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.0786516853932584,
|
| 265 |
+
"grad_norm": 0.394658925096467,
|
| 266 |
+
"learning_rate": 1.1167294115304992e-05,
|
| 267 |
+
"loss": 0.6749,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 2.134831460674157,
|
| 272 |
+
"grad_norm": 0.39767908125309154,
|
| 273 |
+
"learning_rate": 1.0749076892861227e-05,
|
| 274 |
+
"loss": 0.6544,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 2.191011235955056,
|
| 279 |
+
"grad_norm": 0.42673180505937613,
|
| 280 |
+
"learning_rate": 1.0331401507649868e-05,
|
| 281 |
+
"loss": 0.6896,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 2.247191011235955,
|
| 286 |
+
"grad_norm": 0.408239645067392,
|
| 287 |
+
"learning_rate": 9.915169877702096e-06,
|
| 288 |
+
"loss": 0.726,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 2.303370786516854,
|
| 293 |
+
"grad_norm": 0.39464002130518294,
|
| 294 |
+
"learning_rate": 9.501280803439204e-06,
|
| 295 |
+
"loss": 0.6655,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 2.359550561797753,
|
| 300 |
+
"grad_norm": 0.40978323940865435,
|
| 301 |
+
"learning_rate": 9.090628026824941e-06,
|
| 302 |
+
"loss": 0.6764,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 2.4157303370786516,
|
| 307 |
+
"grad_norm": 0.38292295583739444,
|
| 308 |
+
"learning_rate": 8.684098301440903e-06,
|
| 309 |
+
"loss": 0.6719,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 2.4719101123595504,
|
| 314 |
+
"grad_norm": 0.38693445253625824,
|
| 315 |
+
"learning_rate": 8.2825694776525e-06,
|
| 316 |
+
"loss": 0.6584,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.5280898876404496,
|
| 321 |
+
"grad_norm": 0.3994341136157858,
|
| 322 |
+
"learning_rate": 7.886908607000321e-06,
|
| 323 |
+
"loss": 0.6971,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.5842696629213484,
|
| 328 |
+
"grad_norm": 0.40129558631958756,
|
| 329 |
+
"learning_rate": 7.497970069910192e-06,
|
| 330 |
+
"loss": 0.6747,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.640449438202247,
|
| 335 |
+
"grad_norm": 0.39014270480667235,
|
| 336 |
+
"learning_rate": 7.116593730764929e-06,
|
| 337 |
+
"loss": 0.6796,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.696629213483146,
|
| 342 |
+
"grad_norm": 0.36998791500930744,
|
| 343 |
+
"learning_rate": 6.743603124321712e-06,
|
| 344 |
+
"loss": 0.6902,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.752808988764045,
|
| 349 |
+
"grad_norm": 0.3970482683621501,
|
| 350 |
+
"learning_rate": 6.379803677391223e-06,
|
| 351 |
+
"loss": 0.6934,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.808988764044944,
|
| 356 |
+
"grad_norm": 0.3861644580574831,
|
| 357 |
+
"learning_rate": 6.02598096961865e-06,
|
| 358 |
+
"loss": 0.6614,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.865168539325843,
|
| 363 |
+
"grad_norm": 0.3941319315860406,
|
| 364 |
+
"learning_rate": 5.682899037122178e-06,
|
| 365 |
+
"loss": 0.6839,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 2.9213483146067416,
|
| 370 |
+
"grad_norm": 0.3898881332551341,
|
| 371 |
+
"learning_rate": 5.351298722652064e-06,
|
| 372 |
+
"loss": 0.6667,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 2.9775280898876404,
|
| 377 |
+
"grad_norm": 0.40076378703515086,
|
| 378 |
+
"learning_rate": 5.031896075832846e-06,
|
| 379 |
+
"loss": 0.6708,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 3.033707865168539,
|
| 384 |
+
"grad_norm": 0.41791394545484606,
|
| 385 |
+
"learning_rate": 4.725380806943299e-06,
|
| 386 |
+
"loss": 0.6575,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.0898876404494384,
|
| 391 |
+
"grad_norm": 0.41375267265638194,
|
| 392 |
+
"learning_rate": 4.432414797572894e-06,
|
| 393 |
+
"loss": 0.6318,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 3.146067415730337,
|
| 398 |
+
"grad_norm": 0.4218475144549854,
|
| 399 |
+
"learning_rate": 4.153630671370821e-06,
|
| 400 |
+
"loss": 0.6342,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 3.202247191011236,
|
| 405 |
+
"grad_norm": 0.4164065437650473,
|
| 406 |
+
"learning_rate": 3.889630427973951e-06,
|
| 407 |
+
"loss": 0.6274,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 3.258426966292135,
|
| 412 |
+
"grad_norm": 0.41742766148238786,
|
| 413 |
+
"learning_rate": 3.6409841430635166e-06,
|
| 414 |
+
"loss": 0.6078,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 3.3146067415730336,
|
| 419 |
+
"grad_norm": 0.4472934678743176,
|
| 420 |
+
"learning_rate": 3.408228737357575e-06,
|
| 421 |
+
"loss": 0.6137,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 3.370786516853933,
|
| 426 |
+
"grad_norm": 0.3983175026213215,
|
| 427 |
+
"learning_rate": 3.191866817197539e-06,
|
| 428 |
+
"loss": 0.6268,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 3.4269662921348316,
|
| 433 |
+
"grad_norm": 0.41447625127416166,
|
| 434 |
+
"learning_rate": 2.9923655892323144e-06,
|
| 435 |
+
"loss": 0.635,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 3.4831460674157304,
|
| 440 |
+
"grad_norm": 0.3967828745027063,
|
| 441 |
+
"learning_rate": 2.8101558515436506e-06,
|
| 442 |
+
"loss": 0.6181,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 3.539325842696629,
|
| 447 |
+
"grad_norm": 0.39144720476319333,
|
| 448 |
+
"learning_rate": 2.645631063391285e-06,
|
| 449 |
+
"loss": 0.6002,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 3.595505617977528,
|
| 454 |
+
"grad_norm": 0.40314596605224257,
|
| 455 |
+
"learning_rate": 2.4991464955866314e-06,
|
| 456 |
+
"loss": 0.6046,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 3.6516853932584272,
|
| 461 |
+
"grad_norm": 0.3979575371097757,
|
| 462 |
+
"learning_rate": 2.371018463329651e-06,
|
| 463 |
+
"loss": 0.6091,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.7078651685393256,
|
| 468 |
+
"grad_norm": 0.4108961213792489,
|
| 469 |
+
"learning_rate": 2.261523643165532e-06,
|
| 470 |
+
"loss": 0.6445,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 3.764044943820225,
|
| 475 |
+
"grad_norm": 0.40810633042253025,
|
| 476 |
+
"learning_rate": 2.1708984755361205e-06,
|
| 477 |
+
"loss": 0.6311,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 3.8202247191011236,
|
| 482 |
+
"grad_norm": 0.39820144430518867,
|
| 483 |
+
"learning_rate": 2.0993386542161944e-06,
|
| 484 |
+
"loss": 0.6154,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 3.8764044943820224,
|
| 489 |
+
"grad_norm": 0.4070293846963812,
|
| 490 |
+
"learning_rate": 2.0469987037371005e-06,
|
| 491 |
+
"loss": 0.6193,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 3.932584269662921,
|
| 496 |
+
"grad_norm": 0.40297827384808366,
|
| 497 |
+
"learning_rate": 2.013991645710262e-06,
|
| 498 |
+
"loss": 0.6471,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 3.98876404494382,
|
| 503 |
+
"grad_norm": 0.4066161050267969,
|
| 504 |
+
"learning_rate": 2.0003887547710647e-06,
|
| 505 |
+
"loss": 0.6469,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 4.0,
|
| 510 |
+
"step": 356,
|
| 511 |
+
"total_flos": 20283251490816.0,
|
| 512 |
+
"train_loss": 0.7583397939968645,
|
| 513 |
+
"train_runtime": 1935.5651,
|
| 514 |
+
"train_samples_per_second": 11.753,
|
| 515 |
+
"train_steps_per_second": 0.184
|
| 516 |
+
}
|
| 517 |
+
],
|
| 518 |
+
"logging_steps": 5,
|
| 519 |
+
"max_steps": 356,
|
| 520 |
+
"num_input_tokens_seen": 0,
|
| 521 |
+
"num_train_epochs": 4,
|
| 522 |
+
"save_steps": 200,
|
| 523 |
+
"stateful_callbacks": {
|
| 524 |
+
"TrainerControl": {
|
| 525 |
+
"args": {
|
| 526 |
+
"should_epoch_stop": false,
|
| 527 |
+
"should_evaluate": false,
|
| 528 |
+
"should_log": false,
|
| 529 |
+
"should_save": true,
|
| 530 |
+
"should_training_stop": true
|
| 531 |
+
},
|
| 532 |
+
"attributes": {}
|
| 533 |
+
}
|
| 534 |
+
},
|
| 535 |
+
"total_flos": 20283251490816.0,
|
| 536 |
+
"train_batch_size": 8,
|
| 537 |
+
"trial_name": null,
|
| 538 |
+
"trial_params": null
|
| 539 |
+
}
|