Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +511 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Meta-Llama/Meta-Llama-3.1-8B
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: llama-3-1-8b-math-orca-qlora-10k-ep1-frac-100-seed-432
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for llama-3-1-8b-math-orca-qlora-10k-ep1-frac-100-seed-432
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Meta-Llama/Meta-Llama-3.1-8B](https://huggingface.co/Meta-Llama/Meta-Llama-3.1-8B).
|
| 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="kxnguyen/llama-3-1-8b-math-orca-qlora-10k-ep1-frac-100-seed-432", 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 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.1
|
| 37 |
+
- Transformers: 4.49.0
|
| 38 |
+
- Pytorch: 2.5.1
|
| 39 |
+
- Datasets: 3.1.0
|
| 40 |
+
- Tokenizers: 0.21.1
|
| 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": 0.9985228951255539,
|
| 3 |
+
"total_flos": 2.6821312997071258e+17,
|
| 4 |
+
"train_loss": 0.3007896387365443,
|
| 5 |
+
"train_runtime": 3313.8561,
|
| 6 |
+
"train_samples": 10000,
|
| 7 |
+
"train_samples_per_second": 1.632,
|
| 8 |
+
"train_steps_per_second": 0.102
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9985228951255539,
|
| 3 |
+
"total_flos": 2.6821312997071258e+17,
|
| 4 |
+
"train_loss": 0.3007896387365443,
|
| 5 |
+
"train_runtime": 3313.8561,
|
| 6 |
+
"train_samples": 10000,
|
| 7 |
+
"train_samples_per_second": 1.632,
|
| 8 |
+
"train_steps_per_second": 0.102
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,511 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9985228951255539,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 338,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.014771048744460856,
|
| 13 |
+
"grad_norm": 2.5938093662261963,
|
| 14 |
+
"learning_rate": 0.0002,
|
| 15 |
+
"loss": 0.9809,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.029542097488921712,
|
| 20 |
+
"grad_norm": 0.9637813568115234,
|
| 21 |
+
"learning_rate": 0.0002,
|
| 22 |
+
"loss": 0.4769,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04431314623338257,
|
| 27 |
+
"grad_norm": 1.666123628616333,
|
| 28 |
+
"learning_rate": 0.0002,
|
| 29 |
+
"loss": 0.4077,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.059084194977843424,
|
| 34 |
+
"grad_norm": 0.5881379246711731,
|
| 35 |
+
"learning_rate": 0.0002,
|
| 36 |
+
"loss": 0.4035,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07385524372230429,
|
| 41 |
+
"grad_norm": 0.5366687774658203,
|
| 42 |
+
"learning_rate": 0.0002,
|
| 43 |
+
"loss": 0.3695,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08862629246676514,
|
| 48 |
+
"grad_norm": 0.5429549217224121,
|
| 49 |
+
"learning_rate": 0.0002,
|
| 50 |
+
"loss": 0.3579,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.103397341211226,
|
| 55 |
+
"grad_norm": 0.5302016735076904,
|
| 56 |
+
"learning_rate": 0.0002,
|
| 57 |
+
"loss": 0.3623,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.11816838995568685,
|
| 62 |
+
"grad_norm": 0.4706576466560364,
|
| 63 |
+
"learning_rate": 0.0002,
|
| 64 |
+
"loss": 0.3423,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1329394387001477,
|
| 69 |
+
"grad_norm": 0.47507619857788086,
|
| 70 |
+
"learning_rate": 0.0002,
|
| 71 |
+
"loss": 0.3301,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.14771048744460857,
|
| 76 |
+
"grad_norm": 0.487821102142334,
|
| 77 |
+
"learning_rate": 0.0002,
|
| 78 |
+
"loss": 0.3262,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.16248153618906944,
|
| 83 |
+
"grad_norm": 0.46188947558403015,
|
| 84 |
+
"learning_rate": 0.0002,
|
| 85 |
+
"loss": 0.3381,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.17725258493353027,
|
| 90 |
+
"grad_norm": 0.49672871828079224,
|
| 91 |
+
"learning_rate": 0.0002,
|
| 92 |
+
"loss": 0.3474,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.19202363367799113,
|
| 97 |
+
"grad_norm": 0.45688968896865845,
|
| 98 |
+
"learning_rate": 0.0002,
|
| 99 |
+
"loss": 0.3356,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.206794682422452,
|
| 104 |
+
"grad_norm": 0.5083580017089844,
|
| 105 |
+
"learning_rate": 0.0002,
|
| 106 |
+
"loss": 0.317,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.22156573116691286,
|
| 111 |
+
"grad_norm": 0.4326242506504059,
|
| 112 |
+
"learning_rate": 0.0002,
|
| 113 |
+
"loss": 0.3107,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2363367799113737,
|
| 118 |
+
"grad_norm": 0.7657620906829834,
|
| 119 |
+
"learning_rate": 0.0002,
|
| 120 |
+
"loss": 0.3055,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2511078286558346,
|
| 125 |
+
"grad_norm": 0.4073372483253479,
|
| 126 |
+
"learning_rate": 0.0002,
|
| 127 |
+
"loss": 0.3041,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2658788774002954,
|
| 132 |
+
"grad_norm": 0.4194050431251526,
|
| 133 |
+
"learning_rate": 0.0002,
|
| 134 |
+
"loss": 0.3121,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.28064992614475626,
|
| 139 |
+
"grad_norm": 0.4937780499458313,
|
| 140 |
+
"learning_rate": 0.0002,
|
| 141 |
+
"loss": 0.3065,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.29542097488921715,
|
| 146 |
+
"grad_norm": 0.39246585965156555,
|
| 147 |
+
"learning_rate": 0.0002,
|
| 148 |
+
"loss": 0.3081,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.310192023633678,
|
| 153 |
+
"grad_norm": 0.4153652787208557,
|
| 154 |
+
"learning_rate": 0.0002,
|
| 155 |
+
"loss": 0.3074,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.3249630723781389,
|
| 160 |
+
"grad_norm": 0.39885184168815613,
|
| 161 |
+
"learning_rate": 0.0002,
|
| 162 |
+
"loss": 0.3016,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3397341211225997,
|
| 167 |
+
"grad_norm": 0.3999512195587158,
|
| 168 |
+
"learning_rate": 0.0002,
|
| 169 |
+
"loss": 0.302,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.35450516986706054,
|
| 174 |
+
"grad_norm": 0.40937578678131104,
|
| 175 |
+
"learning_rate": 0.0002,
|
| 176 |
+
"loss": 0.2964,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.36927621861152143,
|
| 181 |
+
"grad_norm": 1.0849940776824951,
|
| 182 |
+
"learning_rate": 0.0002,
|
| 183 |
+
"loss": 0.3098,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.38404726735598227,
|
| 188 |
+
"grad_norm": 0.36466699838638306,
|
| 189 |
+
"learning_rate": 0.0002,
|
| 190 |
+
"loss": 0.2964,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3988183161004431,
|
| 195 |
+
"grad_norm": 0.32518795132637024,
|
| 196 |
+
"learning_rate": 0.0002,
|
| 197 |
+
"loss": 0.2788,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.413589364844904,
|
| 202 |
+
"grad_norm": 0.3508060872554779,
|
| 203 |
+
"learning_rate": 0.0002,
|
| 204 |
+
"loss": 0.2758,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.42836041358936483,
|
| 209 |
+
"grad_norm": 0.34023162722587585,
|
| 210 |
+
"learning_rate": 0.0002,
|
| 211 |
+
"loss": 0.2955,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4431314623338257,
|
| 216 |
+
"grad_norm": 0.3429297208786011,
|
| 217 |
+
"learning_rate": 0.0002,
|
| 218 |
+
"loss": 0.2812,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.45790251107828656,
|
| 223 |
+
"grad_norm": 0.3394342064857483,
|
| 224 |
+
"learning_rate": 0.0002,
|
| 225 |
+
"loss": 0.2751,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4726735598227474,
|
| 230 |
+
"grad_norm": 0.3172396421432495,
|
| 231 |
+
"learning_rate": 0.0002,
|
| 232 |
+
"loss": 0.2813,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4874446085672083,
|
| 237 |
+
"grad_norm": 0.5636305809020996,
|
| 238 |
+
"learning_rate": 0.0002,
|
| 239 |
+
"loss": 0.2714,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5022156573116692,
|
| 244 |
+
"grad_norm": 0.33329370617866516,
|
| 245 |
+
"learning_rate": 0.0002,
|
| 246 |
+
"loss": 0.2759,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.51698670605613,
|
| 251 |
+
"grad_norm": 0.34862470626831055,
|
| 252 |
+
"learning_rate": 0.0002,
|
| 253 |
+
"loss": 0.2875,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5317577548005908,
|
| 258 |
+
"grad_norm": 0.41521379351615906,
|
| 259 |
+
"learning_rate": 0.0002,
|
| 260 |
+
"loss": 0.2744,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.5465288035450517,
|
| 265 |
+
"grad_norm": 0.3359523117542267,
|
| 266 |
+
"learning_rate": 0.0002,
|
| 267 |
+
"loss": 0.282,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.5612998522895125,
|
| 272 |
+
"grad_norm": 0.3089170455932617,
|
| 273 |
+
"learning_rate": 0.0002,
|
| 274 |
+
"loss": 0.2628,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5760709010339734,
|
| 279 |
+
"grad_norm": 0.36551329493522644,
|
| 280 |
+
"learning_rate": 0.0002,
|
| 281 |
+
"loss": 0.2776,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5908419497784343,
|
| 286 |
+
"grad_norm": 0.32992231845855713,
|
| 287 |
+
"learning_rate": 0.0002,
|
| 288 |
+
"loss": 0.2599,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6056129985228951,
|
| 293 |
+
"grad_norm": 0.3119284510612488,
|
| 294 |
+
"learning_rate": 0.0002,
|
| 295 |
+
"loss": 0.2699,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.620384047267356,
|
| 300 |
+
"grad_norm": 0.2953311800956726,
|
| 301 |
+
"learning_rate": 0.0002,
|
| 302 |
+
"loss": 0.2705,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.6351550960118169,
|
| 307 |
+
"grad_norm": 0.3757329285144806,
|
| 308 |
+
"learning_rate": 0.0002,
|
| 309 |
+
"loss": 0.2918,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.6499261447562777,
|
| 314 |
+
"grad_norm": 0.36705055832862854,
|
| 315 |
+
"learning_rate": 0.0002,
|
| 316 |
+
"loss": 0.2545,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6646971935007385,
|
| 321 |
+
"grad_norm": 0.3092058002948761,
|
| 322 |
+
"learning_rate": 0.0002,
|
| 323 |
+
"loss": 0.2624,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6794682422451994,
|
| 328 |
+
"grad_norm": 0.31742286682128906,
|
| 329 |
+
"learning_rate": 0.0002,
|
| 330 |
+
"loss": 0.2602,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6942392909896603,
|
| 335 |
+
"grad_norm": 0.2955617308616638,
|
| 336 |
+
"learning_rate": 0.0002,
|
| 337 |
+
"loss": 0.256,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.7090103397341211,
|
| 342 |
+
"grad_norm": 0.3345969617366791,
|
| 343 |
+
"learning_rate": 0.0002,
|
| 344 |
+
"loss": 0.2687,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.723781388478582,
|
| 349 |
+
"grad_norm": 0.2796613276004791,
|
| 350 |
+
"learning_rate": 0.0002,
|
| 351 |
+
"loss": 0.2526,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.7385524372230429,
|
| 356 |
+
"grad_norm": 0.5415365695953369,
|
| 357 |
+
"learning_rate": 0.0002,
|
| 358 |
+
"loss": 0.2545,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.7533234859675036,
|
| 363 |
+
"grad_norm": 0.3844436705112457,
|
| 364 |
+
"learning_rate": 0.0002,
|
| 365 |
+
"loss": 0.2599,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.7680945347119645,
|
| 370 |
+
"grad_norm": 0.3186696171760559,
|
| 371 |
+
"learning_rate": 0.0002,
|
| 372 |
+
"loss": 0.2477,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.7828655834564254,
|
| 377 |
+
"grad_norm": 0.38170936703681946,
|
| 378 |
+
"learning_rate": 0.0002,
|
| 379 |
+
"loss": 0.2582,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.7976366322008862,
|
| 384 |
+
"grad_norm": 0.29369300603866577,
|
| 385 |
+
"learning_rate": 0.0002,
|
| 386 |
+
"loss": 0.2505,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.8124076809453471,
|
| 391 |
+
"grad_norm": 0.29856300354003906,
|
| 392 |
+
"learning_rate": 0.0002,
|
| 393 |
+
"loss": 0.2675,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.827178729689808,
|
| 398 |
+
"grad_norm": 0.2721855342388153,
|
| 399 |
+
"learning_rate": 0.0002,
|
| 400 |
+
"loss": 0.2489,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.8419497784342689,
|
| 405 |
+
"grad_norm": 0.3029973804950714,
|
| 406 |
+
"learning_rate": 0.0002,
|
| 407 |
+
"loss": 0.2575,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.8567208271787297,
|
| 412 |
+
"grad_norm": 0.2983309030532837,
|
| 413 |
+
"learning_rate": 0.0002,
|
| 414 |
+
"loss": 0.2628,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.8714918759231906,
|
| 419 |
+
"grad_norm": 0.5093730092048645,
|
| 420 |
+
"learning_rate": 0.0002,
|
| 421 |
+
"loss": 0.2552,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8862629246676514,
|
| 426 |
+
"grad_norm": 0.28230157494544983,
|
| 427 |
+
"learning_rate": 0.0002,
|
| 428 |
+
"loss": 0.2592,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.9010339734121122,
|
| 433 |
+
"grad_norm": 0.371902197599411,
|
| 434 |
+
"learning_rate": 0.0002,
|
| 435 |
+
"loss": 0.2596,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.9158050221565731,
|
| 440 |
+
"grad_norm": 0.3786104619503021,
|
| 441 |
+
"learning_rate": 0.0002,
|
| 442 |
+
"loss": 0.25,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.930576070901034,
|
| 447 |
+
"grad_norm": 0.4518865942955017,
|
| 448 |
+
"learning_rate": 0.0002,
|
| 449 |
+
"loss": 0.2546,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.9453471196454948,
|
| 454 |
+
"grad_norm": 0.29951682686805725,
|
| 455 |
+
"learning_rate": 0.0002,
|
| 456 |
+
"loss": 0.2433,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.9601181683899557,
|
| 461 |
+
"grad_norm": 0.2999703884124756,
|
| 462 |
+
"learning_rate": 0.0002,
|
| 463 |
+
"loss": 0.2419,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.9748892171344166,
|
| 468 |
+
"grad_norm": 0.2904799282550812,
|
| 469 |
+
"learning_rate": 0.0002,
|
| 470 |
+
"loss": 0.2474,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.9896602658788775,
|
| 475 |
+
"grad_norm": 0.28127652406692505,
|
| 476 |
+
"learning_rate": 0.0002,
|
| 477 |
+
"loss": 0.2458,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.9985228951255539,
|
| 482 |
+
"step": 338,
|
| 483 |
+
"total_flos": 2.6821312997071258e+17,
|
| 484 |
+
"train_loss": 0.3007896387365443,
|
| 485 |
+
"train_runtime": 3313.8561,
|
| 486 |
+
"train_samples_per_second": 1.632,
|
| 487 |
+
"train_steps_per_second": 0.102
|
| 488 |
+
}
|
| 489 |
+
],
|
| 490 |
+
"logging_steps": 5,
|
| 491 |
+
"max_steps": 338,
|
| 492 |
+
"num_input_tokens_seen": 0,
|
| 493 |
+
"num_train_epochs": 1,
|
| 494 |
+
"save_steps": 500,
|
| 495 |
+
"stateful_callbacks": {
|
| 496 |
+
"TrainerControl": {
|
| 497 |
+
"args": {
|
| 498 |
+
"should_epoch_stop": false,
|
| 499 |
+
"should_evaluate": false,
|
| 500 |
+
"should_log": false,
|
| 501 |
+
"should_save": true,
|
| 502 |
+
"should_training_stop": true
|
| 503 |
+
},
|
| 504 |
+
"attributes": {}
|
| 505 |
+
}
|
| 506 |
+
},
|
| 507 |
+
"total_flos": 2.6821312997071258e+17,
|
| 508 |
+
"train_batch_size": 8,
|
| 509 |
+
"trial_name": null,
|
| 510 |
+
"trial_params": null
|
| 511 |
+
}
|