Model save
Browse files- README.md +57 -0
- adapter_model.safetensors +1 -1
- all_results.json +10 -0
- train_results.json +10 -0
- trainer_state.json +1452 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: legis-qwen_2-5-7b-valid
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for legis-qwen_2-5-7b-valid
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-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="felipeoes/legis-qwen_2-5-7b-valid", 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/felipealumni-usp/huggingface/runs/5yl999nv)
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.1
|
| 37 |
+
- Transformers: 4.46.3
|
| 38 |
+
- Pytorch: 2.4.1
|
| 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 |
+
```
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2260784280
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94c466df93b599c03df6cb64767e3fc4e8fa1eaa7044183192e88125e559aa46
|
| 3 |
size 2260784280
|
all_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_samples": 117,
|
| 4 |
+
"total_flos": 2.65944394463445e+17,
|
| 5 |
+
"train_loss": 0.5110978669325849,
|
| 6 |
+
"train_runtime": 64922.9003,
|
| 7 |
+
"train_samples": 116076,
|
| 8 |
+
"train_samples_per_second": 0.29,
|
| 9 |
+
"train_steps_per_second": 0.036
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_samples": 117,
|
| 4 |
+
"total_flos": 2.65944394463445e+17,
|
| 5 |
+
"train_loss": 0.5110978669325849,
|
| 6 |
+
"train_runtime": 64922.9003,
|
| 7 |
+
"train_samples": 116076,
|
| 8 |
+
"train_samples_per_second": 0.29,
|
| 9 |
+
"train_steps_per_second": 0.036
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1452 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4453,
|
| 3 |
+
"best_model_checkpoint": "runs/legis-qwen_2-5-7b-valid/checkpoint-2000",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 25,
|
| 6 |
+
"global_step": 2353,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.010624734381640459,
|
| 13 |
+
"grad_norm": 1.3824252776738029,
|
| 14 |
+
"learning_rate": 2.1186440677966103e-05,
|
| 15 |
+
"loss": 0.9908,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.010624734381640459,
|
| 20 |
+
"eval_loss": 0.8797302842140198,
|
| 21 |
+
"eval_runtime": 13.696,
|
| 22 |
+
"eval_samples_per_second": 1.314,
|
| 23 |
+
"eval_steps_per_second": 0.657,
|
| 24 |
+
"step": 25
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.021249468763280918,
|
| 28 |
+
"grad_norm": 0.5848728162132136,
|
| 29 |
+
"learning_rate": 4.2372881355932206e-05,
|
| 30 |
+
"loss": 0.7379,
|
| 31 |
+
"step": 50
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.021249468763280918,
|
| 35 |
+
"eval_loss": 0.7652693390846252,
|
| 36 |
+
"eval_runtime": 13.4778,
|
| 37 |
+
"eval_samples_per_second": 1.336,
|
| 38 |
+
"eval_steps_per_second": 0.668,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.03187420314492138,
|
| 43 |
+
"grad_norm": 0.4083536495349737,
|
| 44 |
+
"learning_rate": 6.35593220338983e-05,
|
| 45 |
+
"loss": 0.6739,
|
| 46 |
+
"step": 75
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.03187420314492138,
|
| 50 |
+
"eval_loss": 0.7201825976371765,
|
| 51 |
+
"eval_runtime": 13.6633,
|
| 52 |
+
"eval_samples_per_second": 1.317,
|
| 53 |
+
"eval_steps_per_second": 0.659,
|
| 54 |
+
"step": 75
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.042498937526561836,
|
| 58 |
+
"grad_norm": 0.40319269494868826,
|
| 59 |
+
"learning_rate": 8.474576271186441e-05,
|
| 60 |
+
"loss": 0.6333,
|
| 61 |
+
"step": 100
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.042498937526561836,
|
| 65 |
+
"eval_loss": 0.6928180456161499,
|
| 66 |
+
"eval_runtime": 13.5603,
|
| 67 |
+
"eval_samples_per_second": 1.327,
|
| 68 |
+
"eval_steps_per_second": 0.664,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.053123671908202295,
|
| 73 |
+
"grad_norm": 0.45146499021091163,
|
| 74 |
+
"learning_rate": 0.00010593220338983052,
|
| 75 |
+
"loss": 0.6138,
|
| 76 |
+
"step": 125
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.053123671908202295,
|
| 80 |
+
"eval_loss": 0.6759499311447144,
|
| 81 |
+
"eval_runtime": 13.7103,
|
| 82 |
+
"eval_samples_per_second": 1.313,
|
| 83 |
+
"eval_steps_per_second": 0.656,
|
| 84 |
+
"step": 125
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.06374840628984275,
|
| 88 |
+
"grad_norm": 0.4438422289624626,
|
| 89 |
+
"learning_rate": 0.0001271186440677966,
|
| 90 |
+
"loss": 0.6053,
|
| 91 |
+
"step": 150
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.06374840628984275,
|
| 95 |
+
"eval_loss": 0.6635024547576904,
|
| 96 |
+
"eval_runtime": 13.5819,
|
| 97 |
+
"eval_samples_per_second": 1.325,
|
| 98 |
+
"eval_steps_per_second": 0.663,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.0743731406714832,
|
| 103 |
+
"grad_norm": 0.4272726064498105,
|
| 104 |
+
"learning_rate": 0.00014830508474576273,
|
| 105 |
+
"loss": 0.5893,
|
| 106 |
+
"step": 175
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.0743731406714832,
|
| 110 |
+
"eval_loss": 0.6542551517486572,
|
| 111 |
+
"eval_runtime": 13.5533,
|
| 112 |
+
"eval_samples_per_second": 1.328,
|
| 113 |
+
"eval_steps_per_second": 0.664,
|
| 114 |
+
"step": 175
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.08499787505312367,
|
| 118 |
+
"grad_norm": 0.40810966129419957,
|
| 119 |
+
"learning_rate": 0.00016949152542372882,
|
| 120 |
+
"loss": 0.5912,
|
| 121 |
+
"step": 200
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08499787505312367,
|
| 125 |
+
"eval_loss": 0.6452566385269165,
|
| 126 |
+
"eval_runtime": 13.556,
|
| 127 |
+
"eval_samples_per_second": 1.328,
|
| 128 |
+
"eval_steps_per_second": 0.664,
|
| 129 |
+
"step": 200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.09562260943476412,
|
| 133 |
+
"grad_norm": 0.317957905950078,
|
| 134 |
+
"learning_rate": 0.00019067796610169492,
|
| 135 |
+
"loss": 0.588,
|
| 136 |
+
"step": 225
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.09562260943476412,
|
| 140 |
+
"eval_loss": 0.6399930119514465,
|
| 141 |
+
"eval_runtime": 13.7279,
|
| 142 |
+
"eval_samples_per_second": 1.311,
|
| 143 |
+
"eval_steps_per_second": 0.656,
|
| 144 |
+
"step": 225
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.10624734381640459,
|
| 148 |
+
"grad_norm": 0.33193624291905227,
|
| 149 |
+
"learning_rate": 0.00019997841915310352,
|
| 150 |
+
"loss": 0.5738,
|
| 151 |
+
"step": 250
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.10624734381640459,
|
| 155 |
+
"eval_loss": 0.635059654712677,
|
| 156 |
+
"eval_runtime": 13.7296,
|
| 157 |
+
"eval_samples_per_second": 1.311,
|
| 158 |
+
"eval_steps_per_second": 0.656,
|
| 159 |
+
"step": 250
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.11687207819804504,
|
| 163 |
+
"grad_norm": 0.2478181869132683,
|
| 164 |
+
"learning_rate": 0.0001998325689426475,
|
| 165 |
+
"loss": 0.5511,
|
| 166 |
+
"step": 275
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.11687207819804504,
|
| 170 |
+
"eval_loss": 0.6286189556121826,
|
| 171 |
+
"eval_runtime": 13.6848,
|
| 172 |
+
"eval_samples_per_second": 1.315,
|
| 173 |
+
"eval_steps_per_second": 0.658,
|
| 174 |
+
"step": 275
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.1274968125796855,
|
| 178 |
+
"grad_norm": 0.29097194553722133,
|
| 179 |
+
"learning_rate": 0.00019954932703763562,
|
| 180 |
+
"loss": 0.5663,
|
| 181 |
+
"step": 300
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.1274968125796855,
|
| 185 |
+
"eval_loss": 0.6214559078216553,
|
| 186 |
+
"eval_runtime": 13.6708,
|
| 187 |
+
"eval_samples_per_second": 1.317,
|
| 188 |
+
"eval_steps_per_second": 0.658,
|
| 189 |
+
"step": 300
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.13812154696132597,
|
| 193 |
+
"grad_norm": 0.2608796359844066,
|
| 194 |
+
"learning_rate": 0.00019912908324157293,
|
| 195 |
+
"loss": 0.5544,
|
| 196 |
+
"step": 325
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.13812154696132597,
|
| 200 |
+
"eval_loss": 0.615837812423706,
|
| 201 |
+
"eval_runtime": 13.627,
|
| 202 |
+
"eval_samples_per_second": 1.321,
|
| 203 |
+
"eval_steps_per_second": 0.66,
|
| 204 |
+
"step": 325
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.1487462813429664,
|
| 208 |
+
"grad_norm": 0.28941490682658677,
|
| 209 |
+
"learning_rate": 0.00019857241590286702,
|
| 210 |
+
"loss": 0.5553,
|
| 211 |
+
"step": 350
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.1487462813429664,
|
| 215 |
+
"eval_loss": 0.6113396286964417,
|
| 216 |
+
"eval_runtime": 13.5958,
|
| 217 |
+
"eval_samples_per_second": 1.324,
|
| 218 |
+
"eval_steps_per_second": 0.662,
|
| 219 |
+
"step": 350
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.15937101572460688,
|
| 223 |
+
"grad_norm": 0.30035371415468154,
|
| 224 |
+
"learning_rate": 0.00019788009111889248,
|
| 225 |
+
"loss": 0.5665,
|
| 226 |
+
"step": 375
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.15937101572460688,
|
| 230 |
+
"eval_loss": 0.6078417301177979,
|
| 231 |
+
"eval_runtime": 13.54,
|
| 232 |
+
"eval_samples_per_second": 1.329,
|
| 233 |
+
"eval_steps_per_second": 0.665,
|
| 234 |
+
"step": 375
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.16999575010624735,
|
| 238 |
+
"grad_norm": 0.2835386050077782,
|
| 239 |
+
"learning_rate": 0.0001970530616816716,
|
| 240 |
+
"loss": 0.5429,
|
| 241 |
+
"step": 400
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.16999575010624735,
|
| 245 |
+
"eval_loss": 0.6027636528015137,
|
| 246 |
+
"eval_runtime": 13.735,
|
| 247 |
+
"eval_samples_per_second": 1.311,
|
| 248 |
+
"eval_steps_per_second": 0.655,
|
| 249 |
+
"step": 400
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.1806204844878878,
|
| 253 |
+
"grad_norm": 0.22659224554756538,
|
| 254 |
+
"learning_rate": 0.00019609246576662184,
|
| 255 |
+
"loss": 0.5532,
|
| 256 |
+
"step": 425
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.1806204844878878,
|
| 260 |
+
"eval_loss": 0.5958096981048584,
|
| 261 |
+
"eval_runtime": 13.671,
|
| 262 |
+
"eval_samples_per_second": 1.317,
|
| 263 |
+
"eval_steps_per_second": 0.658,
|
| 264 |
+
"step": 425
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.19124521886952825,
|
| 268 |
+
"grad_norm": 0.1914569343973938,
|
| 269 |
+
"learning_rate": 0.00019499962536617454,
|
| 270 |
+
"loss": 0.5554,
|
| 271 |
+
"step": 450
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.19124521886952825,
|
| 275 |
+
"eval_loss": 0.5921241044998169,
|
| 276 |
+
"eval_runtime": 13.7061,
|
| 277 |
+
"eval_samples_per_second": 1.313,
|
| 278 |
+
"eval_steps_per_second": 0.657,
|
| 279 |
+
"step": 450
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.20186995325116872,
|
| 283 |
+
"grad_norm": 0.23443045180153105,
|
| 284 |
+
"learning_rate": 0.00019377604447042127,
|
| 285 |
+
"loss": 0.5428,
|
| 286 |
+
"step": 475
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.20186995325116872,
|
| 290 |
+
"eval_loss": 0.5899159908294678,
|
| 291 |
+
"eval_runtime": 13.5568,
|
| 292 |
+
"eval_samples_per_second": 1.328,
|
| 293 |
+
"eval_steps_per_second": 0.664,
|
| 294 |
+
"step": 475
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.21249468763280918,
|
| 298 |
+
"grad_norm": 0.2161122041133258,
|
| 299 |
+
"learning_rate": 0.00019242340699729047,
|
| 300 |
+
"loss": 0.5514,
|
| 301 |
+
"step": 500
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.21249468763280918,
|
| 305 |
+
"eval_loss": 0.5880197882652283,
|
| 306 |
+
"eval_runtime": 13.624,
|
| 307 |
+
"eval_samples_per_second": 1.321,
|
| 308 |
+
"eval_steps_per_second": 0.661,
|
| 309 |
+
"step": 500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.22311942201444965,
|
| 313 |
+
"grad_norm": 0.2070053292137905,
|
| 314 |
+
"learning_rate": 0.0001909435744751043,
|
| 315 |
+
"loss": 0.5209,
|
| 316 |
+
"step": 525
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.22311942201444965,
|
| 320 |
+
"eval_loss": 0.5879865884780884,
|
| 321 |
+
"eval_runtime": 13.7028,
|
| 322 |
+
"eval_samples_per_second": 1.314,
|
| 323 |
+
"eval_steps_per_second": 0.657,
|
| 324 |
+
"step": 525
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.23374415639609009,
|
| 328 |
+
"grad_norm": 0.23609205770661978,
|
| 329 |
+
"learning_rate": 0.0001893385834807037,
|
| 330 |
+
"loss": 0.5208,
|
| 331 |
+
"step": 550
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.23374415639609009,
|
| 335 |
+
"eval_loss": 0.5853844881057739,
|
| 336 |
+
"eval_runtime": 13.7354,
|
| 337 |
+
"eval_samples_per_second": 1.31,
|
| 338 |
+
"eval_steps_per_second": 0.655,
|
| 339 |
+
"step": 550
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.24436889077773055,
|
| 343 |
+
"grad_norm": 0.19954431376676118,
|
| 344 |
+
"learning_rate": 0.0001876106428366683,
|
| 345 |
+
"loss": 0.5196,
|
| 346 |
+
"step": 575
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.24436889077773055,
|
| 350 |
+
"eval_loss": 0.5810511112213135,
|
| 351 |
+
"eval_runtime": 13.5807,
|
| 352 |
+
"eval_samples_per_second": 1.325,
|
| 353 |
+
"eval_steps_per_second": 0.663,
|
| 354 |
+
"step": 575
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.254993625159371,
|
| 358 |
+
"grad_norm": 0.17831651642966287,
|
| 359 |
+
"learning_rate": 0.00018576213057148815,
|
| 360 |
+
"loss": 0.5278,
|
| 361 |
+
"step": 600
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.254993625159371,
|
| 365 |
+
"eval_loss": 0.5783408880233765,
|
| 366 |
+
"eval_runtime": 13.6189,
|
| 367 |
+
"eval_samples_per_second": 1.322,
|
| 368 |
+
"eval_steps_per_second": 0.661,
|
| 369 |
+
"step": 600
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.2656183595410115,
|
| 373 |
+
"grad_norm": 0.2049029334189066,
|
| 374 |
+
"learning_rate": 0.00018379559064687047,
|
| 375 |
+
"loss": 0.5408,
|
| 376 |
+
"step": 625
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.2656183595410115,
|
| 380 |
+
"eval_loss": 0.575043261051178,
|
| 381 |
+
"eval_runtime": 13.6594,
|
| 382 |
+
"eval_samples_per_second": 1.318,
|
| 383 |
+
"eval_steps_per_second": 0.659,
|
| 384 |
+
"step": 625
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.27624309392265195,
|
| 388 |
+
"grad_norm": 0.1904505136620864,
|
| 389 |
+
"learning_rate": 0.0001817137294566856,
|
| 390 |
+
"loss": 0.5275,
|
| 391 |
+
"step": 650
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.27624309392265195,
|
| 395 |
+
"eval_loss": 0.5728871822357178,
|
| 396 |
+
"eval_runtime": 13.7391,
|
| 397 |
+
"eval_samples_per_second": 1.31,
|
| 398 |
+
"eval_steps_per_second": 0.655,
|
| 399 |
+
"step": 650
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.2868678283042924,
|
| 403 |
+
"grad_norm": 0.22154830832529648,
|
| 404 |
+
"learning_rate": 0.00017951941210237073,
|
| 405 |
+
"loss": 0.533,
|
| 406 |
+
"step": 675
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.2868678283042924,
|
| 410 |
+
"eval_loss": 0.5692403316497803,
|
| 411 |
+
"eval_runtime": 13.545,
|
| 412 |
+
"eval_samples_per_second": 1.329,
|
| 413 |
+
"eval_steps_per_second": 0.664,
|
| 414 |
+
"step": 675
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.2974925626859328,
|
| 418 |
+
"grad_norm": 0.2174115807792657,
|
| 419 |
+
"learning_rate": 0.00017721565844991643,
|
| 420 |
+
"loss": 0.5258,
|
| 421 |
+
"step": 700
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.2974925626859328,
|
| 425 |
+
"eval_loss": 0.5680803656578064,
|
| 426 |
+
"eval_runtime": 13.7424,
|
| 427 |
+
"eval_samples_per_second": 1.31,
|
| 428 |
+
"eval_steps_per_second": 0.655,
|
| 429 |
+
"step": 700
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.3081172970675733,
|
| 433 |
+
"grad_norm": 0.19207104639056627,
|
| 434 |
+
"learning_rate": 0.00017480563897386326,
|
| 435 |
+
"loss": 0.5175,
|
| 436 |
+
"step": 725
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3081172970675733,
|
| 440 |
+
"eval_loss": 0.5652768611907959,
|
| 441 |
+
"eval_runtime": 13.5314,
|
| 442 |
+
"eval_samples_per_second": 1.33,
|
| 443 |
+
"eval_steps_per_second": 0.665,
|
| 444 |
+
"step": 725
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.31874203144921376,
|
| 448 |
+
"grad_norm": 0.19862950596297196,
|
| 449 |
+
"learning_rate": 0.00017229267039402753,
|
| 450 |
+
"loss": 0.5053,
|
| 451 |
+
"step": 750
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.31874203144921376,
|
| 455 |
+
"eval_loss": 0.5653724074363708,
|
| 456 |
+
"eval_runtime": 13.5197,
|
| 457 |
+
"eval_samples_per_second": 1.331,
|
| 458 |
+
"eval_steps_per_second": 0.666,
|
| 459 |
+
"step": 750
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.3293667658308542,
|
| 463 |
+
"grad_norm": 0.2636122695391369,
|
| 464 |
+
"learning_rate": 0.00016968021111096133,
|
| 465 |
+
"loss": 0.5215,
|
| 466 |
+
"step": 775
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.3293667658308542,
|
| 470 |
+
"eval_loss": 0.5632820129394531,
|
| 471 |
+
"eval_runtime": 13.7575,
|
| 472 |
+
"eval_samples_per_second": 1.308,
|
| 473 |
+
"eval_steps_per_second": 0.654,
|
| 474 |
+
"step": 775
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.3399915002124947,
|
| 478 |
+
"grad_norm": 0.17073712293046736,
|
| 479 |
+
"learning_rate": 0.00016697185644642853,
|
| 480 |
+
"loss": 0.5205,
|
| 481 |
+
"step": 800
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.3399915002124947,
|
| 485 |
+
"eval_loss": 0.5606775879859924,
|
| 486 |
+
"eval_runtime": 13.6853,
|
| 487 |
+
"eval_samples_per_second": 1.315,
|
| 488 |
+
"eval_steps_per_second": 0.658,
|
| 489 |
+
"step": 800
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.35061623459413516,
|
| 493 |
+
"grad_norm": 0.1877562077800468,
|
| 494 |
+
"learning_rate": 0.00016417133369544677,
|
| 495 |
+
"loss": 0.5026,
|
| 496 |
+
"step": 825
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.35061623459413516,
|
| 500 |
+
"eval_loss": 0.5574628710746765,
|
| 501 |
+
"eval_runtime": 13.6524,
|
| 502 |
+
"eval_samples_per_second": 1.318,
|
| 503 |
+
"eval_steps_per_second": 0.659,
|
| 504 |
+
"step": 825
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.3612409689757756,
|
| 508 |
+
"grad_norm": 0.2187753211287567,
|
| 509 |
+
"learning_rate": 0.0001612824969967052,
|
| 510 |
+
"loss": 0.5219,
|
| 511 |
+
"step": 850
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.3612409689757756,
|
| 515 |
+
"eval_loss": 0.5507112145423889,
|
| 516 |
+
"eval_runtime": 13.5887,
|
| 517 |
+
"eval_samples_per_second": 1.325,
|
| 518 |
+
"eval_steps_per_second": 0.662,
|
| 519 |
+
"step": 850
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.3718657033574161,
|
| 523 |
+
"grad_norm": 0.21822616884068516,
|
| 524 |
+
"learning_rate": 0.00015830932202841723,
|
| 525 |
+
"loss": 0.5138,
|
| 526 |
+
"step": 875
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.3718657033574161,
|
| 530 |
+
"eval_loss": 0.5505152940750122,
|
| 531 |
+
"eval_runtime": 13.6074,
|
| 532 |
+
"eval_samples_per_second": 1.323,
|
| 533 |
+
"eval_steps_per_second": 0.661,
|
| 534 |
+
"step": 875
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3824904377390565,
|
| 538 |
+
"grad_norm": 0.19921946854007808,
|
| 539 |
+
"learning_rate": 0.00015525590053690804,
|
| 540 |
+
"loss": 0.5082,
|
| 541 |
+
"step": 900
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3824904377390565,
|
| 545 |
+
"eval_loss": 0.5506541728973389,
|
| 546 |
+
"eval_runtime": 13.8251,
|
| 547 |
+
"eval_samples_per_second": 1.302,
|
| 548 |
+
"eval_steps_per_second": 0.651,
|
| 549 |
+
"step": 900
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.39311517212069697,
|
| 553 |
+
"grad_norm": 0.22351938753382847,
|
| 554 |
+
"learning_rate": 0.00015212643470546668,
|
| 555 |
+
"loss": 0.5089,
|
| 556 |
+
"step": 925
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.39311517212069697,
|
| 560 |
+
"eval_loss": 0.548447847366333,
|
| 561 |
+
"eval_runtime": 13.6917,
|
| 562 |
+
"eval_samples_per_second": 1.315,
|
| 563 |
+
"eval_steps_per_second": 0.657,
|
| 564 |
+
"step": 925
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.40373990650233743,
|
| 568 |
+
"grad_norm": 0.19131472458469853,
|
| 569 |
+
"learning_rate": 0.00014892523137121245,
|
| 570 |
+
"loss": 0.5241,
|
| 571 |
+
"step": 950
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.40373990650233743,
|
| 575 |
+
"eval_loss": 0.5444693565368652,
|
| 576 |
+
"eval_runtime": 13.7071,
|
| 577 |
+
"eval_samples_per_second": 1.313,
|
| 578 |
+
"eval_steps_per_second": 0.657,
|
| 579 |
+
"step": 950
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.4143646408839779,
|
| 583 |
+
"grad_norm": 0.19421007360537948,
|
| 584 |
+
"learning_rate": 0.00014565669609793442,
|
| 585 |
+
"loss": 0.507,
|
| 586 |
+
"step": 975
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.4143646408839779,
|
| 590 |
+
"eval_loss": 0.5411471724510193,
|
| 591 |
+
"eval_runtime": 13.6537,
|
| 592 |
+
"eval_samples_per_second": 1.318,
|
| 593 |
+
"eval_steps_per_second": 0.659,
|
| 594 |
+
"step": 975
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.42498937526561836,
|
| 598 |
+
"grad_norm": 0.19516131766473133,
|
| 599 |
+
"learning_rate": 0.00014232532711306155,
|
| 600 |
+
"loss": 0.4931,
|
| 601 |
+
"step": 1000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.42498937526561836,
|
| 605 |
+
"eval_loss": 0.5411571860313416,
|
| 606 |
+
"eval_runtime": 13.7576,
|
| 607 |
+
"eval_samples_per_second": 1.308,
|
| 608 |
+
"eval_steps_per_second": 0.654,
|
| 609 |
+
"step": 1000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.43561410964725883,
|
| 613 |
+
"grad_norm": 0.16870156967259758,
|
| 614 |
+
"learning_rate": 0.00013893570911710656,
|
| 615 |
+
"loss": 0.5033,
|
| 616 |
+
"step": 1025
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.43561410964725883,
|
| 620 |
+
"eval_loss": 0.5401651859283447,
|
| 621 |
+
"eval_runtime": 13.575,
|
| 622 |
+
"eval_samples_per_second": 1.326,
|
| 623 |
+
"eval_steps_per_second": 0.663,
|
| 624 |
+
"step": 1025
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.4462388440288993,
|
| 628 |
+
"grad_norm": 0.1637560423009355,
|
| 629 |
+
"learning_rate": 0.00013549250697410427,
|
| 630 |
+
"loss": 0.4781,
|
| 631 |
+
"step": 1050
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.4462388440288993,
|
| 635 |
+
"eval_loss": 0.5378057956695557,
|
| 636 |
+
"eval_runtime": 13.6055,
|
| 637 |
+
"eval_samples_per_second": 1.323,
|
| 638 |
+
"eval_steps_per_second": 0.661,
|
| 639 |
+
"step": 1050
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.45686357841053976,
|
| 643 |
+
"grad_norm": 0.17989029128302533,
|
| 644 |
+
"learning_rate": 0.00013200045929172679,
|
| 645 |
+
"loss": 0.4962,
|
| 646 |
+
"step": 1075
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.45686357841053976,
|
| 650 |
+
"eval_loss": 0.5310940742492676,
|
| 651 |
+
"eval_runtime": 13.7133,
|
| 652 |
+
"eval_samples_per_second": 1.313,
|
| 653 |
+
"eval_steps_per_second": 0.656,
|
| 654 |
+
"step": 1075
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.46748831279218017,
|
| 658 |
+
"grad_norm": 0.18360206121827216,
|
| 659 |
+
"learning_rate": 0.00012846437189991162,
|
| 660 |
+
"loss": 0.5189,
|
| 661 |
+
"step": 1100
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.46748831279218017,
|
| 665 |
+
"eval_loss": 0.5309380292892456,
|
| 666 |
+
"eval_runtime": 13.6608,
|
| 667 |
+
"eval_samples_per_second": 1.318,
|
| 668 |
+
"eval_steps_per_second": 0.659,
|
| 669 |
+
"step": 1100
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.47811304717382064,
|
| 673 |
+
"grad_norm": 0.18474682238955895,
|
| 674 |
+
"learning_rate": 0.00012488911123697643,
|
| 675 |
+
"loss": 0.5085,
|
| 676 |
+
"step": 1125
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.47811304717382064,
|
| 680 |
+
"eval_loss": 0.5284863114356995,
|
| 681 |
+
"eval_runtime": 13.6774,
|
| 682 |
+
"eval_samples_per_second": 1.316,
|
| 683 |
+
"eval_steps_per_second": 0.658,
|
| 684 |
+
"step": 1125
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.4887377815554611,
|
| 688 |
+
"grad_norm": 0.16088542652738752,
|
| 689 |
+
"learning_rate": 0.00012127959765232386,
|
| 690 |
+
"loss": 0.4958,
|
| 691 |
+
"step": 1150
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.4887377815554611,
|
| 695 |
+
"eval_loss": 0.5277774930000305,
|
| 696 |
+
"eval_runtime": 13.5397,
|
| 697 |
+
"eval_samples_per_second": 1.329,
|
| 698 |
+
"eval_steps_per_second": 0.665,
|
| 699 |
+
"step": 1150
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.49936251593710157,
|
| 703 |
+
"grad_norm": 0.16824774903806683,
|
| 704 |
+
"learning_rate": 0.00011764079863495239,
|
| 705 |
+
"loss": 0.4772,
|
| 706 |
+
"step": 1175
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.49936251593710157,
|
| 710 |
+
"eval_loss": 0.5244553089141846,
|
| 711 |
+
"eval_runtime": 13.57,
|
| 712 |
+
"eval_samples_per_second": 1.326,
|
| 713 |
+
"eval_steps_per_second": 0.663,
|
| 714 |
+
"step": 1175
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.509987250318742,
|
| 718 |
+
"grad_norm": 0.19032915808439513,
|
| 719 |
+
"learning_rate": 0.00011397772197709304,
|
| 720 |
+
"loss": 0.4951,
|
| 721 |
+
"step": 1200
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.509987250318742,
|
| 725 |
+
"eval_loss": 0.5219236612319946,
|
| 726 |
+
"eval_runtime": 13.751,
|
| 727 |
+
"eval_samples_per_second": 1.309,
|
| 728 |
+
"eval_steps_per_second": 0.654,
|
| 729 |
+
"step": 1200
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.5206119847003825,
|
| 733 |
+
"grad_norm": 0.19732678264253936,
|
| 734 |
+
"learning_rate": 0.00011029540888237941,
|
| 735 |
+
"loss": 0.4598,
|
| 736 |
+
"step": 1225
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.5206119847003825,
|
| 740 |
+
"eval_loss": 0.5182512402534485,
|
| 741 |
+
"eval_runtime": 13.5188,
|
| 742 |
+
"eval_samples_per_second": 1.331,
|
| 743 |
+
"eval_steps_per_second": 0.666,
|
| 744 |
+
"step": 1225
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.531236719082023,
|
| 748 |
+
"grad_norm": 0.18322879510839657,
|
| 749 |
+
"learning_rate": 0.00010659892702803677,
|
| 750 |
+
"loss": 0.4937,
|
| 751 |
+
"step": 1250
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.531236719082023,
|
| 755 |
+
"eval_loss": 0.5168578624725342,
|
| 756 |
+
"eval_runtime": 13.6434,
|
| 757 |
+
"eval_samples_per_second": 1.319,
|
| 758 |
+
"eval_steps_per_second": 0.66,
|
| 759 |
+
"step": 1250
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.5418614534636634,
|
| 763 |
+
"grad_norm": 0.16469463785666816,
|
| 764 |
+
"learning_rate": 0.00010289336359063726,
|
| 765 |
+
"loss": 0.4972,
|
| 766 |
+
"step": 1275
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.5418614534636634,
|
| 770 |
+
"eval_loss": 0.515425443649292,
|
| 771 |
+
"eval_runtime": 13.5093,
|
| 772 |
+
"eval_samples_per_second": 1.332,
|
| 773 |
+
"eval_steps_per_second": 0.666,
|
| 774 |
+
"step": 1275
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.5524861878453039,
|
| 778 |
+
"grad_norm": 0.18400238161361088,
|
| 779 |
+
"learning_rate": 9.918381824501988e-05,
|
| 780 |
+
"loss": 0.4761,
|
| 781 |
+
"step": 1300
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.5524861878453039,
|
| 785 |
+
"eval_loss": 0.5136659145355225,
|
| 786 |
+
"eval_runtime": 13.5018,
|
| 787 |
+
"eval_samples_per_second": 1.333,
|
| 788 |
+
"eval_steps_per_second": 0.667,
|
| 789 |
+
"step": 1300
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.5631109222269444,
|
| 793 |
+
"grad_norm": 0.17535283435867968,
|
| 794 |
+
"learning_rate": 9.547539614601e-05,
|
| 795 |
+
"loss": 0.4888,
|
| 796 |
+
"step": 1325
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.5631109222269444,
|
| 800 |
+
"eval_loss": 0.509550929069519,
|
| 801 |
+
"eval_runtime": 13.7492,
|
| 802 |
+
"eval_samples_per_second": 1.309,
|
| 803 |
+
"eval_steps_per_second": 0.655,
|
| 804 |
+
"step": 1325
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.5737356566085848,
|
| 808 |
+
"grad_norm": 0.19027736106694362,
|
| 809 |
+
"learning_rate": 9.17732009025972e-05,
|
| 810 |
+
"loss": 0.4788,
|
| 811 |
+
"step": 1350
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.5737356566085848,
|
| 815 |
+
"eval_loss": 0.5091434717178345,
|
| 816 |
+
"eval_runtime": 13.6692,
|
| 817 |
+
"eval_samples_per_second": 1.317,
|
| 818 |
+
"eval_steps_per_second": 0.658,
|
| 819 |
+
"step": 1350
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.5843603909902253,
|
| 823 |
+
"grad_norm": 0.14945795860859679,
|
| 824 |
+
"learning_rate": 8.808232755424066e-05,
|
| 825 |
+
"loss": 0.4711,
|
| 826 |
+
"step": 1375
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.5843603909902253,
|
| 830 |
+
"eval_loss": 0.5075156688690186,
|
| 831 |
+
"eval_runtime": 13.5509,
|
| 832 |
+
"eval_samples_per_second": 1.328,
|
| 833 |
+
"eval_steps_per_second": 0.664,
|
| 834 |
+
"step": 1375
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.5949851253718657,
|
| 838 |
+
"grad_norm": 0.1876115407072482,
|
| 839 |
+
"learning_rate": 8.440785555896786e-05,
|
| 840 |
+
"loss": 0.4841,
|
| 841 |
+
"step": 1400
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.5949851253718657,
|
| 845 |
+
"eval_loss": 0.505312442779541,
|
| 846 |
+
"eval_runtime": 13.6058,
|
| 847 |
+
"eval_samples_per_second": 1.323,
|
| 848 |
+
"eval_steps_per_second": 0.661,
|
| 849 |
+
"step": 1400
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6056098597535061,
|
| 853 |
+
"grad_norm": 0.17126471384455877,
|
| 854 |
+
"learning_rate": 8.075484180291701e-05,
|
| 855 |
+
"loss": 0.4733,
|
| 856 |
+
"step": 1425
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6056098597535061,
|
| 860 |
+
"eval_loss": 0.5053431391716003,
|
| 861 |
+
"eval_runtime": 13.6732,
|
| 862 |
+
"eval_samples_per_second": 1.316,
|
| 863 |
+
"eval_steps_per_second": 0.658,
|
| 864 |
+
"step": 1425
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.6162345941351466,
|
| 868 |
+
"grad_norm": 0.1612132057853942,
|
| 869 |
+
"learning_rate": 7.712831364094332e-05,
|
| 870 |
+
"loss": 0.4801,
|
| 871 |
+
"step": 1450
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.6162345941351466,
|
| 875 |
+
"eval_loss": 0.5035179853439331,
|
| 876 |
+
"eval_runtime": 13.5528,
|
| 877 |
+
"eval_samples_per_second": 1.328,
|
| 878 |
+
"eval_steps_per_second": 0.664,
|
| 879 |
+
"step": 1450
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.626859328516787,
|
| 883 |
+
"grad_norm": 0.1774699324298755,
|
| 884 |
+
"learning_rate": 7.353326197786673e-05,
|
| 885 |
+
"loss": 0.4735,
|
| 886 |
+
"step": 1475
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.626859328516787,
|
| 890 |
+
"eval_loss": 0.5026558041572571,
|
| 891 |
+
"eval_runtime": 13.6414,
|
| 892 |
+
"eval_samples_per_second": 1.32,
|
| 893 |
+
"eval_steps_per_second": 0.66,
|
| 894 |
+
"step": 1475
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.6374840628984275,
|
| 898 |
+
"grad_norm": 0.15538798680363947,
|
| 899 |
+
"learning_rate": 6.997463439988324e-05,
|
| 900 |
+
"loss": 0.4677,
|
| 901 |
+
"step": 1500
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.6374840628984275,
|
| 905 |
+
"eval_loss": 0.5013994574546814,
|
| 906 |
+
"eval_runtime": 13.5624,
|
| 907 |
+
"eval_samples_per_second": 1.327,
|
| 908 |
+
"eval_steps_per_second": 0.664,
|
| 909 |
+
"step": 1500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.648108797280068,
|
| 913 |
+
"grad_norm": 0.1819125971720679,
|
| 914 |
+
"learning_rate": 6.645732836559187e-05,
|
| 915 |
+
"loss": 0.4588,
|
| 916 |
+
"step": 1525
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.648108797280068,
|
| 920 |
+
"eval_loss": 0.5008442997932434,
|
| 921 |
+
"eval_runtime": 13.7199,
|
| 922 |
+
"eval_samples_per_second": 1.312,
|
| 923 |
+
"eval_steps_per_second": 0.656,
|
| 924 |
+
"step": 1525
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 0.6587335316617084,
|
| 928 |
+
"grad_norm": 0.17692677070822024,
|
| 929 |
+
"learning_rate": 6.298618446600856e-05,
|
| 930 |
+
"loss": 0.4803,
|
| 931 |
+
"step": 1550
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.6587335316617084,
|
| 935 |
+
"eval_loss": 0.5003005266189575,
|
| 936 |
+
"eval_runtime": 13.6645,
|
| 937 |
+
"eval_samples_per_second": 1.317,
|
| 938 |
+
"eval_steps_per_second": 0.659,
|
| 939 |
+
"step": 1550
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.6693582660433489,
|
| 943 |
+
"grad_norm": 0.19223708797121006,
|
| 944 |
+
"learning_rate": 5.956597976284239e-05,
|
| 945 |
+
"loss": 0.4801,
|
| 946 |
+
"step": 1575
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.6693582660433489,
|
| 950 |
+
"eval_loss": 0.49848344922065735,
|
| 951 |
+
"eval_runtime": 13.7844,
|
| 952 |
+
"eval_samples_per_second": 1.306,
|
| 953 |
+
"eval_steps_per_second": 0.653,
|
| 954 |
+
"step": 1575
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.6799830004249894,
|
| 958 |
+
"grad_norm": 0.19050620001754145,
|
| 959 |
+
"learning_rate": 5.620142121420201e-05,
|
| 960 |
+
"loss": 0.4699,
|
| 961 |
+
"step": 1600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6799830004249894,
|
| 965 |
+
"eval_loss": 0.4974481761455536,
|
| 966 |
+
"eval_runtime": 13.6881,
|
| 967 |
+
"eval_samples_per_second": 1.315,
|
| 968 |
+
"eval_steps_per_second": 0.658,
|
| 969 |
+
"step": 1600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.6906077348066298,
|
| 973 |
+
"grad_norm": 0.18849962999043668,
|
| 974 |
+
"learning_rate": 5.289713919678034e-05,
|
| 975 |
+
"loss": 0.4746,
|
| 976 |
+
"step": 1625
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.6906077348066298,
|
| 980 |
+
"eval_loss": 0.49532318115234375,
|
| 981 |
+
"eval_runtime": 13.6989,
|
| 982 |
+
"eval_samples_per_second": 1.314,
|
| 983 |
+
"eval_steps_per_second": 0.657,
|
| 984 |
+
"step": 1625
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.7012324691882703,
|
| 988 |
+
"grad_norm": 0.17255206587971877,
|
| 989 |
+
"learning_rate": 4.965768113343233e-05,
|
| 990 |
+
"loss": 0.4677,
|
| 991 |
+
"step": 1650
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.7012324691882703,
|
| 995 |
+
"eval_loss": 0.4947270154953003,
|
| 996 |
+
"eval_runtime": 13.6686,
|
| 997 |
+
"eval_samples_per_second": 1.317,
|
| 998 |
+
"eval_steps_per_second": 0.658,
|
| 999 |
+
"step": 1650
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.7118572035699108,
|
| 1003 |
+
"grad_norm": 0.1772053939786707,
|
| 1004 |
+
"learning_rate": 4.6487505234914876e-05,
|
| 1005 |
+
"loss": 0.4673,
|
| 1006 |
+
"step": 1675
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.7118572035699108,
|
| 1010 |
+
"eval_loss": 0.49380868673324585,
|
| 1011 |
+
"eval_runtime": 13.5528,
|
| 1012 |
+
"eval_samples_per_second": 1.328,
|
| 1013 |
+
"eval_steps_per_second": 0.664,
|
| 1014 |
+
"step": 1675
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.7224819379515512,
|
| 1018 |
+
"grad_norm": 0.17206430964699912,
|
| 1019 |
+
"learning_rate": 4.339097436440346e-05,
|
| 1020 |
+
"loss": 0.4794,
|
| 1021 |
+
"step": 1700
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.7224819379515512,
|
| 1025 |
+
"eval_loss": 0.4924837648868561,
|
| 1026 |
+
"eval_runtime": 13.6884,
|
| 1027 |
+
"eval_samples_per_second": 1.315,
|
| 1028 |
+
"eval_steps_per_second": 0.657,
|
| 1029 |
+
"step": 1700
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.7331066723331917,
|
| 1033 |
+
"grad_norm": 0.20781990513713502,
|
| 1034 |
+
"learning_rate": 4.037235003322685e-05,
|
| 1035 |
+
"loss": 0.4688,
|
| 1036 |
+
"step": 1725
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.7331066723331917,
|
| 1040 |
+
"eval_loss": 0.49092790484428406,
|
| 1041 |
+
"eval_runtime": 13.5314,
|
| 1042 |
+
"eval_samples_per_second": 1.33,
|
| 1043 |
+
"eval_steps_per_second": 0.665,
|
| 1044 |
+
"step": 1725
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.7437314067148322,
|
| 1048 |
+
"grad_norm": 0.15566309201486278,
|
| 1049 |
+
"learning_rate": 3.7435786536085573e-05,
|
| 1050 |
+
"loss": 0.4789,
|
| 1051 |
+
"step": 1750
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.7437314067148322,
|
| 1055 |
+
"eval_loss": 0.48968595266342163,
|
| 1056 |
+
"eval_runtime": 13.561,
|
| 1057 |
+
"eval_samples_per_second": 1.327,
|
| 1058 |
+
"eval_steps_per_second": 0.664,
|
| 1059 |
+
"step": 1750
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7543561410964726,
|
| 1063 |
+
"grad_norm": 0.23381879921923462,
|
| 1064 |
+
"learning_rate": 3.458532523382323e-05,
|
| 1065 |
+
"loss": 0.4826,
|
| 1066 |
+
"step": 1775
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7543561410964726,
|
| 1070 |
+
"eval_loss": 0.4890773594379425,
|
| 1071 |
+
"eval_runtime": 13.5661,
|
| 1072 |
+
"eval_samples_per_second": 1.327,
|
| 1073 |
+
"eval_steps_per_second": 0.663,
|
| 1074 |
+
"step": 1775
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.764980875478113,
|
| 1078 |
+
"grad_norm": 0.16721926504642332,
|
| 1079 |
+
"learning_rate": 3.1824888991620294e-05,
|
| 1080 |
+
"loss": 0.4642,
|
| 1081 |
+
"step": 1800
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.764980875478113,
|
| 1085 |
+
"eval_loss": 0.4890780448913574,
|
| 1086 |
+
"eval_runtime": 13.719,
|
| 1087 |
+
"eval_samples_per_second": 1.312,
|
| 1088 |
+
"eval_steps_per_second": 0.656,
|
| 1089 |
+
"step": 1800
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.7756056098597535,
|
| 1093 |
+
"grad_norm": 0.18751577560760466,
|
| 1094 |
+
"learning_rate": 2.9158276780264006e-05,
|
| 1095 |
+
"loss": 0.4613,
|
| 1096 |
+
"step": 1825
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.7756056098597535,
|
| 1100 |
+
"eval_loss": 0.48852983117103577,
|
| 1101 |
+
"eval_runtime": 13.741,
|
| 1102 |
+
"eval_samples_per_second": 1.31,
|
| 1103 |
+
"eval_steps_per_second": 0.655,
|
| 1104 |
+
"step": 1825
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.7862303442413939,
|
| 1108 |
+
"grad_norm": 0.1728653190685702,
|
| 1109 |
+
"learning_rate": 2.6589158447924454e-05,
|
| 1110 |
+
"loss": 0.4615,
|
| 1111 |
+
"step": 1850
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 0.7862303442413939,
|
| 1115 |
+
"eval_loss": 0.4880680441856384,
|
| 1116 |
+
"eval_runtime": 13.6489,
|
| 1117 |
+
"eval_samples_per_second": 1.319,
|
| 1118 |
+
"eval_steps_per_second": 0.659,
|
| 1119 |
+
"step": 1850
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.7968550786230344,
|
| 1123 |
+
"grad_norm": 0.18900325234160106,
|
| 1124 |
+
"learning_rate": 2.4121069669631556e-05,
|
| 1125 |
+
"loss": 0.4587,
|
| 1126 |
+
"step": 1875
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 0.7968550786230344,
|
| 1130 |
+
"eval_loss": 0.4873946011066437,
|
| 1131 |
+
"eval_runtime": 13.7858,
|
| 1132 |
+
"eval_samples_per_second": 1.306,
|
| 1133 |
+
"eval_steps_per_second": 0.653,
|
| 1134 |
+
"step": 1875
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 0.8074798130046749,
|
| 1138 |
+
"grad_norm": 0.14783723259837353,
|
| 1139 |
+
"learning_rate": 2.175740708140468e-05,
|
| 1140 |
+
"loss": 0.4679,
|
| 1141 |
+
"step": 1900
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 0.8074798130046749,
|
| 1145 |
+
"eval_loss": 0.4868377149105072,
|
| 1146 |
+
"eval_runtime": 13.8172,
|
| 1147 |
+
"eval_samples_per_second": 1.303,
|
| 1148 |
+
"eval_steps_per_second": 0.651,
|
| 1149 |
+
"step": 1900
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.8181045473863153,
|
| 1153 |
+
"grad_norm": 0.17087609511278531,
|
| 1154 |
+
"learning_rate": 1.9501423605729773e-05,
|
| 1155 |
+
"loss": 0.4738,
|
| 1156 |
+
"step": 1925
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.8181045473863153,
|
| 1160 |
+
"eval_loss": 0.4864993095397949,
|
| 1161 |
+
"eval_runtime": 13.5968,
|
| 1162 |
+
"eval_samples_per_second": 1.324,
|
| 1163 |
+
"eval_steps_per_second": 0.662,
|
| 1164 |
+
"step": 1925
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.8287292817679558,
|
| 1168 |
+
"grad_norm": 0.18872610979845497,
|
| 1169 |
+
"learning_rate": 1.735622397481901e-05,
|
| 1170 |
+
"loss": 0.4586,
|
| 1171 |
+
"step": 1950
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.8287292817679558,
|
| 1175 |
+
"eval_loss": 0.4859856069087982,
|
| 1176 |
+
"eval_runtime": 13.665,
|
| 1177 |
+
"eval_samples_per_second": 1.317,
|
| 1178 |
+
"eval_steps_per_second": 0.659,
|
| 1179 |
+
"step": 1950
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.8393540161495963,
|
| 1183 |
+
"grad_norm": 0.16502873183660738,
|
| 1184 |
+
"learning_rate": 1.5324760457812302e-05,
|
| 1185 |
+
"loss": 0.4713,
|
| 1186 |
+
"step": 1975
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.8393540161495963,
|
| 1190 |
+
"eval_loss": 0.48532745242118835,
|
| 1191 |
+
"eval_runtime": 13.52,
|
| 1192 |
+
"eval_samples_per_second": 1.331,
|
| 1193 |
+
"eval_steps_per_second": 0.666,
|
| 1194 |
+
"step": 1975
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.8499787505312367,
|
| 1198 |
+
"grad_norm": 0.14410289002259907,
|
| 1199 |
+
"learning_rate": 1.3409828797802205e-05,
|
| 1200 |
+
"loss": 0.4453,
|
| 1201 |
+
"step": 2000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.8499787505312367,
|
| 1205 |
+
"eval_loss": 0.485173761844635,
|
| 1206 |
+
"eval_runtime": 13.584,
|
| 1207 |
+
"eval_samples_per_second": 1.325,
|
| 1208 |
+
"eval_steps_per_second": 0.663,
|
| 1209 |
+
"step": 2000
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.8606034849128772,
|
| 1213 |
+
"grad_norm": 0.1948838377567367,
|
| 1214 |
+
"learning_rate": 1.161406436427297e-05,
|
| 1215 |
+
"loss": 0.473,
|
| 1216 |
+
"step": 2025
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.8606034849128772,
|
| 1220 |
+
"eval_loss": 0.48516422510147095,
|
| 1221 |
+
"eval_runtime": 13.5308,
|
| 1222 |
+
"eval_samples_per_second": 1.33,
|
| 1223 |
+
"eval_steps_per_second": 0.665,
|
| 1224 |
+
"step": 2025
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.8712282192945177,
|
| 1228 |
+
"grad_norm": 0.17372475645367427,
|
| 1229 |
+
"learning_rate": 9.939938526249337e-06,
|
| 1230 |
+
"loss": 0.4499,
|
| 1231 |
+
"step": 2050
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.8712282192945177,
|
| 1235 |
+
"eval_loss": 0.484728068113327,
|
| 1236 |
+
"eval_runtime": 13.4398,
|
| 1237 |
+
"eval_samples_per_second": 1.339,
|
| 1238 |
+
"eval_steps_per_second": 0.67,
|
| 1239 |
+
"step": 2050
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 0.8818529536761581,
|
| 1243 |
+
"grad_norm": 0.15074819772648787,
|
| 1244 |
+
"learning_rate": 8.389755251146159e-06,
|
| 1245 |
+
"loss": 0.4631,
|
| 1246 |
+
"step": 2075
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.8818529536761581,
|
| 1250 |
+
"eval_loss": 0.4841134250164032,
|
| 1251 |
+
"eval_runtime": 13.53,
|
| 1252 |
+
"eval_samples_per_second": 1.33,
|
| 1253 |
+
"eval_steps_per_second": 0.665,
|
| 1254 |
+
"step": 2075
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.8924776880577986,
|
| 1258 |
+
"grad_norm": 0.18378461042583777,
|
| 1259 |
+
"learning_rate": 6.965647933999742e-06,
|
| 1260 |
+
"loss": 0.4671,
|
| 1261 |
+
"step": 2100
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.8924776880577986,
|
| 1265 |
+
"eval_loss": 0.4838300347328186,
|
| 1266 |
+
"eval_runtime": 13.4675,
|
| 1267 |
+
"eval_samples_per_second": 1.337,
|
| 1268 |
+
"eval_steps_per_second": 0.668,
|
| 1269 |
+
"step": 2100
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.9031024224394391,
|
| 1273 |
+
"grad_norm": 0.1564603040061458,
|
| 1274 |
+
"learning_rate": 5.669576461444248e-06,
|
| 1275 |
+
"loss": 0.4463,
|
| 1276 |
+
"step": 2125
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.9031024224394391,
|
| 1280 |
+
"eval_loss": 0.4835617244243622,
|
| 1281 |
+
"eval_runtime": 13.4511,
|
| 1282 |
+
"eval_samples_per_second": 1.338,
|
| 1283 |
+
"eval_steps_per_second": 0.669,
|
| 1284 |
+
"step": 2125
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.9137271568210795,
|
| 1288 |
+
"grad_norm": 0.1814808133368016,
|
| 1289 |
+
"learning_rate": 4.503324514474483e-06,
|
| 1290 |
+
"loss": 0.4778,
|
| 1291 |
+
"step": 2150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.9137271568210795,
|
| 1295 |
+
"eval_loss": 0.48341307044029236,
|
| 1296 |
+
"eval_runtime": 13.483,
|
| 1297 |
+
"eval_samples_per_second": 1.335,
|
| 1298 |
+
"eval_steps_per_second": 0.668,
|
| 1299 |
+
"step": 2150
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.9243518912027199,
|
| 1303 |
+
"grad_norm": 0.16529041233191302,
|
| 1304 |
+
"learning_rate": 3.4684971137061484e-06,
|
| 1305 |
+
"loss": 0.4682,
|
| 1306 |
+
"step": 2175
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.9243518912027199,
|
| 1310 |
+
"eval_loss": 0.4833216667175293,
|
| 1311 |
+
"eval_runtime": 13.6309,
|
| 1312 |
+
"eval_samples_per_second": 1.321,
|
| 1313 |
+
"eval_steps_per_second": 0.66,
|
| 1314 |
+
"step": 2175
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 0.9349766255843603,
|
| 1318 |
+
"grad_norm": 0.16432869665531463,
|
| 1319 |
+
"learning_rate": 2.566518410512586e-06,
|
| 1320 |
+
"loss": 0.4462,
|
| 1321 |
+
"step": 2200
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 0.9349766255843603,
|
| 1325 |
+
"eval_loss": 0.48317214846611023,
|
| 1326 |
+
"eval_runtime": 13.6155,
|
| 1327 |
+
"eval_samples_per_second": 1.322,
|
| 1328 |
+
"eval_steps_per_second": 0.661,
|
| 1329 |
+
"step": 2200
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.9456013599660008,
|
| 1333 |
+
"grad_norm": 0.17027661720242618,
|
| 1334 |
+
"learning_rate": 1.7986297270775054e-06,
|
| 1335 |
+
"loss": 0.4573,
|
| 1336 |
+
"step": 2225
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.9456013599660008,
|
| 1340 |
+
"eval_loss": 0.48307955265045166,
|
| 1341 |
+
"eval_runtime": 13.4168,
|
| 1342 |
+
"eval_samples_per_second": 1.342,
|
| 1343 |
+
"eval_steps_per_second": 0.671,
|
| 1344 |
+
"step": 2225
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.9562260943476413,
|
| 1348 |
+
"grad_norm": 0.16582718568525256,
|
| 1349 |
+
"learning_rate": 1.16588784806112e-06,
|
| 1350 |
+
"loss": 0.4764,
|
| 1351 |
+
"step": 2250
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 0.9562260943476413,
|
| 1355 |
+
"eval_loss": 0.48304203152656555,
|
| 1356 |
+
"eval_runtime": 13.4315,
|
| 1357 |
+
"eval_samples_per_second": 1.34,
|
| 1358 |
+
"eval_steps_per_second": 0.67,
|
| 1359 |
+
"step": 2250
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.9668508287292817,
|
| 1363 |
+
"grad_norm": 0.16474906223134622,
|
| 1364 |
+
"learning_rate": 6.691635662307017e-07,
|
| 1365 |
+
"loss": 0.467,
|
| 1366 |
+
"step": 2275
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.9668508287292817,
|
| 1370 |
+
"eval_loss": 0.4829782247543335,
|
| 1371 |
+
"eval_runtime": 13.5849,
|
| 1372 |
+
"eval_samples_per_second": 1.325,
|
| 1373 |
+
"eval_steps_per_second": 0.663,
|
| 1374 |
+
"step": 2275
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.9774755631109222,
|
| 1378 |
+
"grad_norm": 0.1846923892020628,
|
| 1379 |
+
"learning_rate": 3.091404840571799e-07,
|
| 1380 |
+
"loss": 0.4592,
|
| 1381 |
+
"step": 2300
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.9774755631109222,
|
| 1385 |
+
"eval_loss": 0.4829777479171753,
|
| 1386 |
+
"eval_runtime": 13.5512,
|
| 1387 |
+
"eval_samples_per_second": 1.328,
|
| 1388 |
+
"eval_steps_per_second": 0.664,
|
| 1389 |
+
"step": 2300
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.9881002974925627,
|
| 1393 |
+
"grad_norm": 0.16625664738759183,
|
| 1394 |
+
"learning_rate": 8.631407292692962e-08,
|
| 1395 |
+
"loss": 0.4582,
|
| 1396 |
+
"step": 2325
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.9881002974925627,
|
| 1400 |
+
"eval_loss": 0.48291730880737305,
|
| 1401 |
+
"eval_runtime": 13.3992,
|
| 1402 |
+
"eval_samples_per_second": 1.343,
|
| 1403 |
+
"eval_steps_per_second": 0.672,
|
| 1404 |
+
"step": 2325
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.9987250318742031,
|
| 1408 |
+
"grad_norm": 0.16884372416843668,
|
| 1409 |
+
"learning_rate": 9.909912635497343e-10,
|
| 1410 |
+
"loss": 0.4576,
|
| 1411 |
+
"step": 2350
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.9987250318742031,
|
| 1415 |
+
"eval_loss": 0.4829167127609253,
|
| 1416 |
+
"eval_runtime": 13.4537,
|
| 1417 |
+
"eval_samples_per_second": 1.338,
|
| 1418 |
+
"eval_steps_per_second": 0.669,
|
| 1419 |
+
"step": 2350
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 1.0,
|
| 1423 |
+
"step": 2353,
|
| 1424 |
+
"total_flos": 2.65944394463445e+17,
|
| 1425 |
+
"train_loss": 0.5110978669325849,
|
| 1426 |
+
"train_runtime": 64922.9003,
|
| 1427 |
+
"train_samples_per_second": 0.29,
|
| 1428 |
+
"train_steps_per_second": 0.036
|
| 1429 |
+
}
|
| 1430 |
+
],
|
| 1431 |
+
"logging_steps": 25,
|
| 1432 |
+
"max_steps": 2353,
|
| 1433 |
+
"num_input_tokens_seen": 0,
|
| 1434 |
+
"num_train_epochs": 1,
|
| 1435 |
+
"save_steps": 25,
|
| 1436 |
+
"stateful_callbacks": {
|
| 1437 |
+
"TrainerControl": {
|
| 1438 |
+
"args": {
|
| 1439 |
+
"should_epoch_stop": false,
|
| 1440 |
+
"should_evaluate": false,
|
| 1441 |
+
"should_log": false,
|
| 1442 |
+
"should_save": true,
|
| 1443 |
+
"should_training_stop": true
|
| 1444 |
+
},
|
| 1445 |
+
"attributes": {}
|
| 1446 |
+
}
|
| 1447 |
+
},
|
| 1448 |
+
"total_flos": 2.65944394463445e+17,
|
| 1449 |
+
"train_batch_size": 2,
|
| 1450 |
+
"trial_name": null,
|
| 1451 |
+
"trial_params": null
|
| 1452 |
+
}
|