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 +1302 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: TucanoBR/Tucano-2b4-Instruct
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: legis-tucano-2b-valid
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for legis-tucano-2b-valid
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [TucanoBR/Tucano-2b4-Instruct](https://huggingface.co/TucanoBR/Tucano-2b4-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-tucano-2b-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/iebkq1bq)
|
| 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 370017192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f292d2cbd8a5425f7e3c19b14fe0770dc054ed4e9df88a16e19421366deac93
|
| 3 |
size 370017192
|
all_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9994065281899109,
|
| 3 |
+
"eval_samples": 246,
|
| 4 |
+
"total_flos": 6.615025481416704e+16,
|
| 5 |
+
"train_loss": 0.6960575034103031,
|
| 6 |
+
"train_runtime": 22091.905,
|
| 7 |
+
"train_samples": 245028,
|
| 8 |
+
"train_samples_per_second": 2.44,
|
| 9 |
+
"train_steps_per_second": 0.038
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9994065281899109,
|
| 3 |
+
"eval_samples": 246,
|
| 4 |
+
"total_flos": 6.615025481416704e+16,
|
| 5 |
+
"train_loss": 0.6960575034103031,
|
| 6 |
+
"train_runtime": 22091.905,
|
| 7 |
+
"train_samples": 245028,
|
| 8 |
+
"train_samples_per_second": 2.44,
|
| 9 |
+
"train_steps_per_second": 0.038
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1302 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7003,
|
| 3 |
+
"best_model_checkpoint": "runs/legis-tucano-2b-valid/checkpoint-710",
|
| 4 |
+
"epoch": 0.9994065281899109,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 842,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.011869436201780416,
|
| 13 |
+
"grad_norm": 1.3277966052420052,
|
| 14 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 15 |
+
"loss": 1.3563,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.011869436201780416,
|
| 20 |
+
"eval_loss": 1.2333489656448364,
|
| 21 |
+
"eval_runtime": 7.1731,
|
| 22 |
+
"eval_samples_per_second": 7.249,
|
| 23 |
+
"eval_steps_per_second": 0.976,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02373887240356083,
|
| 28 |
+
"grad_norm": 0.2988625315142482,
|
| 29 |
+
"learning_rate": 4.705882352941177e-05,
|
| 30 |
+
"loss": 1.2538,
|
| 31 |
+
"step": 20
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.02373887240356083,
|
| 35 |
+
"eval_loss": 1.117323398590088,
|
| 36 |
+
"eval_runtime": 7.1487,
|
| 37 |
+
"eval_samples_per_second": 7.274,
|
| 38 |
+
"eval_steps_per_second": 0.979,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.03560830860534125,
|
| 43 |
+
"grad_norm": 0.15105150205016427,
|
| 44 |
+
"learning_rate": 7.058823529411765e-05,
|
| 45 |
+
"loss": 1.1462,
|
| 46 |
+
"step": 30
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.03560830860534125,
|
| 50 |
+
"eval_loss": 1.0231214761734009,
|
| 51 |
+
"eval_runtime": 7.1358,
|
| 52 |
+
"eval_samples_per_second": 7.287,
|
| 53 |
+
"eval_steps_per_second": 0.981,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.04747774480712166,
|
| 58 |
+
"grad_norm": 0.07095151302909826,
|
| 59 |
+
"learning_rate": 9.411764705882353e-05,
|
| 60 |
+
"loss": 1.0635,
|
| 61 |
+
"step": 40
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.04747774480712166,
|
| 65 |
+
"eval_loss": 0.9480358958244324,
|
| 66 |
+
"eval_runtime": 7.1596,
|
| 67 |
+
"eval_samples_per_second": 7.263,
|
| 68 |
+
"eval_steps_per_second": 0.978,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.05934718100890208,
|
| 73 |
+
"grad_norm": 0.0510771961177552,
|
| 74 |
+
"learning_rate": 0.00011764705882352942,
|
| 75 |
+
"loss": 0.9848,
|
| 76 |
+
"step": 50
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.05934718100890208,
|
| 80 |
+
"eval_loss": 0.8925089240074158,
|
| 81 |
+
"eval_runtime": 7.1518,
|
| 82 |
+
"eval_samples_per_second": 7.271,
|
| 83 |
+
"eval_steps_per_second": 0.979,
|
| 84 |
+
"step": 50
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.0712166172106825,
|
| 88 |
+
"grad_norm": 0.040539214461615646,
|
| 89 |
+
"learning_rate": 0.0001411764705882353,
|
| 90 |
+
"loss": 0.9513,
|
| 91 |
+
"step": 60
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.0712166172106825,
|
| 95 |
+
"eval_loss": 0.8546838164329529,
|
| 96 |
+
"eval_runtime": 7.1461,
|
| 97 |
+
"eval_samples_per_second": 7.277,
|
| 98 |
+
"eval_steps_per_second": 0.98,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.0830860534124629,
|
| 103 |
+
"grad_norm": 0.038791558909017074,
|
| 104 |
+
"learning_rate": 0.0001647058823529412,
|
| 105 |
+
"loss": 0.9142,
|
| 106 |
+
"step": 70
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.0830860534124629,
|
| 110 |
+
"eval_loss": 0.8263636231422424,
|
| 111 |
+
"eval_runtime": 7.3085,
|
| 112 |
+
"eval_samples_per_second": 7.115,
|
| 113 |
+
"eval_steps_per_second": 0.958,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.09495548961424333,
|
| 118 |
+
"grad_norm": 0.04169274461251293,
|
| 119 |
+
"learning_rate": 0.00018823529411764707,
|
| 120 |
+
"loss": 0.8854,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.09495548961424333,
|
| 125 |
+
"eval_loss": 0.8020428419113159,
|
| 126 |
+
"eval_runtime": 7.247,
|
| 127 |
+
"eval_samples_per_second": 7.175,
|
| 128 |
+
"eval_steps_per_second": 0.966,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.10682492581602374,
|
| 133 |
+
"grad_norm": 0.043015312738591396,
|
| 134 |
+
"learning_rate": 0.00019997847206287532,
|
| 135 |
+
"loss": 0.8632,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.10682492581602374,
|
| 140 |
+
"eval_loss": 0.7831454873085022,
|
| 141 |
+
"eval_runtime": 7.2244,
|
| 142 |
+
"eval_samples_per_second": 7.198,
|
| 143 |
+
"eval_steps_per_second": 0.969,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.11869436201780416,
|
| 148 |
+
"grad_norm": 0.05736189956437236,
|
| 149 |
+
"learning_rate": 0.00019980630417613612,
|
| 150 |
+
"loss": 0.8669,
|
| 151 |
+
"step": 100
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.11869436201780416,
|
| 155 |
+
"eval_loss": 0.7694521546363831,
|
| 156 |
+
"eval_runtime": 7.2277,
|
| 157 |
+
"eval_samples_per_second": 7.194,
|
| 158 |
+
"eval_steps_per_second": 0.968,
|
| 159 |
+
"step": 100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.13056379821958458,
|
| 163 |
+
"grad_norm": 0.04913014792038086,
|
| 164 |
+
"learning_rate": 0.0001994622648842964,
|
| 165 |
+
"loss": 0.8314,
|
| 166 |
+
"step": 110
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.13056379821958458,
|
| 170 |
+
"eval_loss": 0.7584220170974731,
|
| 171 |
+
"eval_runtime": 7.2488,
|
| 172 |
+
"eval_samples_per_second": 7.174,
|
| 173 |
+
"eval_steps_per_second": 0.966,
|
| 174 |
+
"step": 110
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.142433234421365,
|
| 178 |
+
"grad_norm": 0.04446480334718679,
|
| 179 |
+
"learning_rate": 0.00019894694664007728,
|
| 180 |
+
"loss": 0.8223,
|
| 181 |
+
"step": 120
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.142433234421365,
|
| 185 |
+
"eval_loss": 0.749190092086792,
|
| 186 |
+
"eval_runtime": 7.2447,
|
| 187 |
+
"eval_samples_per_second": 7.178,
|
| 188 |
+
"eval_steps_per_second": 0.966,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.1543026706231454,
|
| 193 |
+
"grad_norm": 0.05216966997945006,
|
| 194 |
+
"learning_rate": 0.00019826123684704952,
|
| 195 |
+
"loss": 0.8141,
|
| 196 |
+
"step": 130
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.1543026706231454,
|
| 200 |
+
"eval_loss": 0.7411435842514038,
|
| 201 |
+
"eval_runtime": 7.2499,
|
| 202 |
+
"eval_samples_per_second": 7.173,
|
| 203 |
+
"eval_steps_per_second": 0.966,
|
| 204 |
+
"step": 130
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.1661721068249258,
|
| 208 |
+
"grad_norm": 0.04233605582577254,
|
| 209 |
+
"learning_rate": 0.00019740631633148045,
|
| 210 |
+
"loss": 0.8023,
|
| 211 |
+
"step": 140
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.1661721068249258,
|
| 215 |
+
"eval_loss": 0.7342764139175415,
|
| 216 |
+
"eval_runtime": 7.2264,
|
| 217 |
+
"eval_samples_per_second": 7.196,
|
| 218 |
+
"eval_steps_per_second": 0.969,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.17804154302670624,
|
| 223 |
+
"grad_norm": 0.04704056707854844,
|
| 224 |
+
"learning_rate": 0.00019638365730889265,
|
| 225 |
+
"loss": 0.7895,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.17804154302670624,
|
| 230 |
+
"eval_loss": 0.7279341220855713,
|
| 231 |
+
"eval_runtime": 7.2397,
|
| 232 |
+
"eval_samples_per_second": 7.183,
|
| 233 |
+
"eval_steps_per_second": 0.967,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.18991097922848665,
|
| 238 |
+
"grad_norm": 0.044026531833322384,
|
| 239 |
+
"learning_rate": 0.00019519502084883582,
|
| 240 |
+
"loss": 0.7916,
|
| 241 |
+
"step": 160
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.18991097922848665,
|
| 245 |
+
"eval_loss": 0.7229446768760681,
|
| 246 |
+
"eval_runtime": 7.2583,
|
| 247 |
+
"eval_samples_per_second": 7.164,
|
| 248 |
+
"eval_steps_per_second": 0.964,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.20178041543026706,
|
| 253 |
+
"grad_norm": 0.05191891103370369,
|
| 254 |
+
"learning_rate": 0.00019384245384223767,
|
| 255 |
+
"loss": 0.791,
|
| 256 |
+
"step": 170
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.20178041543026706,
|
| 260 |
+
"eval_loss": 0.7180152535438538,
|
| 261 |
+
"eval_runtime": 7.2578,
|
| 262 |
+
"eval_samples_per_second": 7.165,
|
| 263 |
+
"eval_steps_per_second": 0.964,
|
| 264 |
+
"step": 170
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.21364985163204747,
|
| 268 |
+
"grad_norm": 0.05572444654938424,
|
| 269 |
+
"learning_rate": 0.00019232828547655615,
|
| 270 |
+
"loss": 0.7955,
|
| 271 |
+
"step": 180
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.21364985163204747,
|
| 275 |
+
"eval_loss": 0.7139907479286194,
|
| 276 |
+
"eval_runtime": 7.2429,
|
| 277 |
+
"eval_samples_per_second": 7.179,
|
| 278 |
+
"eval_steps_per_second": 0.966,
|
| 279 |
+
"step": 180
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.22551928783382788,
|
| 283 |
+
"grad_norm": 0.0516837327891167,
|
| 284 |
+
"learning_rate": 0.00019065512322480332,
|
| 285 |
+
"loss": 0.7776,
|
| 286 |
+
"step": 190
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.22551928783382788,
|
| 290 |
+
"eval_loss": 0.7107158899307251,
|
| 291 |
+
"eval_runtime": 7.2537,
|
| 292 |
+
"eval_samples_per_second": 7.169,
|
| 293 |
+
"eval_steps_per_second": 0.965,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.23738872403560832,
|
| 298 |
+
"grad_norm": 0.050846230869752186,
|
| 299 |
+
"learning_rate": 0.00018882584835534737,
|
| 300 |
+
"loss": 0.785,
|
| 301 |
+
"step": 200
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.23738872403560832,
|
| 305 |
+
"eval_loss": 0.7068641185760498,
|
| 306 |
+
"eval_runtime": 7.2578,
|
| 307 |
+
"eval_samples_per_second": 7.165,
|
| 308 |
+
"eval_steps_per_second": 0.964,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.24925816023738873,
|
| 313 |
+
"grad_norm": 0.05343887989356088,
|
| 314 |
+
"learning_rate": 0.00018684361097022568,
|
| 315 |
+
"loss": 0.7825,
|
| 316 |
+
"step": 210
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.24925816023738873,
|
| 320 |
+
"eval_loss": 0.703826904296875,
|
| 321 |
+
"eval_runtime": 7.2465,
|
| 322 |
+
"eval_samples_per_second": 7.176,
|
| 323 |
+
"eval_steps_per_second": 0.966,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.26112759643916916,
|
| 328 |
+
"grad_norm": 0.05242956764694665,
|
| 329 |
+
"learning_rate": 0.00018471182458051287,
|
| 330 |
+
"loss": 0.7539,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.26112759643916916,
|
| 335 |
+
"eval_loss": 0.7010646462440491,
|
| 336 |
+
"eval_runtime": 7.2455,
|
| 337 |
+
"eval_samples_per_second": 7.177,
|
| 338 |
+
"eval_steps_per_second": 0.966,
|
| 339 |
+
"step": 220
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.27299703264094954,
|
| 343 |
+
"grad_norm": 0.051064873049660696,
|
| 344 |
+
"learning_rate": 0.00018243416022808547,
|
| 345 |
+
"loss": 0.7785,
|
| 346 |
+
"step": 230
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.27299703264094954,
|
| 350 |
+
"eval_loss": 0.6979869604110718,
|
| 351 |
+
"eval_runtime": 7.2598,
|
| 352 |
+
"eval_samples_per_second": 7.163,
|
| 353 |
+
"eval_steps_per_second": 0.964,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.28486646884273,
|
| 358 |
+
"grad_norm": 0.046329966712940596,
|
| 359 |
+
"learning_rate": 0.00018001454016390586,
|
| 360 |
+
"loss": 0.7792,
|
| 361 |
+
"step": 240
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.28486646884273,
|
| 365 |
+
"eval_loss": 0.6954675912857056,
|
| 366 |
+
"eval_runtime": 7.2628,
|
| 367 |
+
"eval_samples_per_second": 7.16,
|
| 368 |
+
"eval_steps_per_second": 0.964,
|
| 369 |
+
"step": 240
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.29673590504451036,
|
| 373 |
+
"grad_norm": 0.05390500282652213,
|
| 374 |
+
"learning_rate": 0.00017745713109371139,
|
| 375 |
+
"loss": 0.753,
|
| 376 |
+
"step": 250
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.29673590504451036,
|
| 380 |
+
"eval_loss": 0.6930329203605652,
|
| 381 |
+
"eval_runtime": 7.2472,
|
| 382 |
+
"eval_samples_per_second": 7.175,
|
| 383 |
+
"eval_steps_per_second": 0.966,
|
| 384 |
+
"step": 250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.3086053412462908,
|
| 388 |
+
"grad_norm": 0.043546536548247354,
|
| 389 |
+
"learning_rate": 0.00017476633700274068,
|
| 390 |
+
"loss": 0.765,
|
| 391 |
+
"step": 260
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.3086053412462908,
|
| 395 |
+
"eval_loss": 0.6907065510749817,
|
| 396 |
+
"eval_runtime": 7.2525,
|
| 397 |
+
"eval_samples_per_second": 7.17,
|
| 398 |
+
"eval_steps_per_second": 0.965,
|
| 399 |
+
"step": 260
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.32047477744807124,
|
| 403 |
+
"grad_norm": 0.06771031218563213,
|
| 404 |
+
"learning_rate": 0.00017194679157185255,
|
| 405 |
+
"loss": 0.7464,
|
| 406 |
+
"step": 270
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.32047477744807124,
|
| 410 |
+
"eval_loss": 0.6883173584938049,
|
| 411 |
+
"eval_runtime": 7.1658,
|
| 412 |
+
"eval_samples_per_second": 7.257,
|
| 413 |
+
"eval_steps_per_second": 0.977,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.3323442136498516,
|
| 418 |
+
"grad_norm": 0.04731592266489432,
|
| 419 |
+
"learning_rate": 0.00016900335019809782,
|
| 420 |
+
"loss": 0.7649,
|
| 421 |
+
"step": 280
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.3323442136498516,
|
| 425 |
+
"eval_loss": 0.6865535378456116,
|
| 426 |
+
"eval_runtime": 7.1591,
|
| 427 |
+
"eval_samples_per_second": 7.264,
|
| 428 |
+
"eval_steps_per_second": 0.978,
|
| 429 |
+
"step": 280
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.34421364985163205,
|
| 433 |
+
"grad_norm": 0.05422567227576833,
|
| 434 |
+
"learning_rate": 0.00016594108163348493,
|
| 435 |
+
"loss": 0.764,
|
| 436 |
+
"step": 290
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.34421364985163205,
|
| 440 |
+
"eval_loss": 0.6839883327484131,
|
| 441 |
+
"eval_runtime": 7.1803,
|
| 442 |
+
"eval_samples_per_second": 7.242,
|
| 443 |
+
"eval_steps_per_second": 0.975,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.3560830860534125,
|
| 448 |
+
"grad_norm": 0.04716111383242718,
|
| 449 |
+
"learning_rate": 0.0001627652592563373,
|
| 450 |
+
"loss": 0.7457,
|
| 451 |
+
"step": 300
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.3560830860534125,
|
| 455 |
+
"eval_loss": 0.6814995408058167,
|
| 456 |
+
"eval_runtime": 7.1545,
|
| 457 |
+
"eval_samples_per_second": 7.268,
|
| 458 |
+
"eval_steps_per_second": 0.978,
|
| 459 |
+
"step": 300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.36795252225519287,
|
| 463 |
+
"grad_norm": 0.04833237070306901,
|
| 464 |
+
"learning_rate": 0.00015948135199027474,
|
| 465 |
+
"loss": 0.7427,
|
| 466 |
+
"step": 310
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.36795252225519287,
|
| 470 |
+
"eval_loss": 0.6795143485069275,
|
| 471 |
+
"eval_runtime": 7.164,
|
| 472 |
+
"eval_samples_per_second": 7.258,
|
| 473 |
+
"eval_steps_per_second": 0.977,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.3798219584569733,
|
| 478 |
+
"grad_norm": 0.04790062682337991,
|
| 479 |
+
"learning_rate": 0.00015609501488645554,
|
| 480 |
+
"loss": 0.7564,
|
| 481 |
+
"step": 320
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.3798219584569733,
|
| 485 |
+
"eval_loss": 0.6774563193321228,
|
| 486 |
+
"eval_runtime": 7.1529,
|
| 487 |
+
"eval_samples_per_second": 7.27,
|
| 488 |
+
"eval_steps_per_second": 0.979,
|
| 489 |
+
"step": 320
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.3916913946587537,
|
| 493 |
+
"grad_norm": 0.04849257718202548,
|
| 494 |
+
"learning_rate": 0.00015261207938529808,
|
| 495 |
+
"loss": 0.7352,
|
| 496 |
+
"step": 330
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.3916913946587537,
|
| 500 |
+
"eval_loss": 0.676394522190094,
|
| 501 |
+
"eval_runtime": 7.1522,
|
| 502 |
+
"eval_samples_per_second": 7.27,
|
| 503 |
+
"eval_steps_per_second": 0.979,
|
| 504 |
+
"step": 330
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.4035608308605341,
|
| 508 |
+
"grad_norm": 0.0449523115474861,
|
| 509 |
+
"learning_rate": 0.00014903854327445116,
|
| 510 |
+
"loss": 0.7471,
|
| 511 |
+
"step": 340
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.4035608308605341,
|
| 515 |
+
"eval_loss": 0.6733577251434326,
|
| 516 |
+
"eval_runtime": 7.1516,
|
| 517 |
+
"eval_samples_per_second": 7.271,
|
| 518 |
+
"eval_steps_per_second": 0.979,
|
| 519 |
+
"step": 340
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.41543026706231456,
|
| 523 |
+
"grad_norm": 0.044399706693004984,
|
| 524 |
+
"learning_rate": 0.00014538056036030622,
|
| 525 |
+
"loss": 0.7408,
|
| 526 |
+
"step": 350
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.41543026706231456,
|
| 530 |
+
"eval_loss": 0.6722955107688904,
|
| 531 |
+
"eval_runtime": 7.1523,
|
| 532 |
+
"eval_samples_per_second": 7.27,
|
| 533 |
+
"eval_steps_per_second": 0.979,
|
| 534 |
+
"step": 350
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.42729970326409494,
|
| 538 |
+
"grad_norm": 0.04749195230072245,
|
| 539 |
+
"learning_rate": 0.00014164442987083762,
|
| 540 |
+
"loss": 0.7493,
|
| 541 |
+
"step": 360
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.42729970326409494,
|
| 545 |
+
"eval_loss": 0.6708287000656128,
|
| 546 |
+
"eval_runtime": 7.158,
|
| 547 |
+
"eval_samples_per_second": 7.265,
|
| 548 |
+
"eval_steps_per_second": 0.978,
|
| 549 |
+
"step": 360
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.4391691394658754,
|
| 553 |
+
"grad_norm": 0.04622607097065686,
|
| 554 |
+
"learning_rate": 0.0001378365856080198,
|
| 555 |
+
"loss": 0.7375,
|
| 556 |
+
"step": 370
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.4391691394658754,
|
| 560 |
+
"eval_loss": 0.6693353056907654,
|
| 561 |
+
"eval_runtime": 7.1529,
|
| 562 |
+
"eval_samples_per_second": 7.27,
|
| 563 |
+
"eval_steps_per_second": 0.979,
|
| 564 |
+
"step": 370
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.45103857566765576,
|
| 568 |
+
"grad_norm": 0.04779103392401475,
|
| 569 |
+
"learning_rate": 0.00013396358486850103,
|
| 570 |
+
"loss": 0.7469,
|
| 571 |
+
"step": 380
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.45103857566765576,
|
| 575 |
+
"eval_loss": 0.6675475835800171,
|
| 576 |
+
"eval_runtime": 7.1609,
|
| 577 |
+
"eval_samples_per_second": 7.262,
|
| 578 |
+
"eval_steps_per_second": 0.978,
|
| 579 |
+
"step": 380
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.4629080118694362,
|
| 583 |
+
"grad_norm": 0.04367335948535579,
|
| 584 |
+
"learning_rate": 0.0001300320971516136,
|
| 585 |
+
"loss": 0.7388,
|
| 586 |
+
"step": 390
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.4629080118694362,
|
| 590 |
+
"eval_loss": 0.6652542948722839,
|
| 591 |
+
"eval_runtime": 7.1535,
|
| 592 |
+
"eval_samples_per_second": 7.269,
|
| 593 |
+
"eval_steps_per_second": 0.979,
|
| 594 |
+
"step": 390
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.47477744807121663,
|
| 598 |
+
"grad_norm": 0.04906513653224005,
|
| 599 |
+
"learning_rate": 0.0001260488926741651,
|
| 600 |
+
"loss": 0.7434,
|
| 601 |
+
"step": 400
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.47477744807121663,
|
| 605 |
+
"eval_loss": 0.6640903949737549,
|
| 606 |
+
"eval_runtime": 7.1588,
|
| 607 |
+
"eval_samples_per_second": 7.264,
|
| 608 |
+
"eval_steps_per_second": 0.978,
|
| 609 |
+
"step": 400
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.486646884272997,
|
| 613 |
+
"grad_norm": 0.04507001139303972,
|
| 614 |
+
"learning_rate": 0.00012202083071178938,
|
| 615 |
+
"loss": 0.7271,
|
| 616 |
+
"step": 410
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.486646884272997,
|
| 620 |
+
"eval_loss": 0.6632689237594604,
|
| 621 |
+
"eval_runtime": 7.1544,
|
| 622 |
+
"eval_samples_per_second": 7.268,
|
| 623 |
+
"eval_steps_per_second": 0.978,
|
| 624 |
+
"step": 410
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.49851632047477745,
|
| 628 |
+
"grad_norm": 0.054746961897722525,
|
| 629 |
+
"learning_rate": 0.00011795484778693382,
|
| 630 |
+
"loss": 0.731,
|
| 631 |
+
"step": 420
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.49851632047477745,
|
| 635 |
+
"eval_loss": 0.6615473031997681,
|
| 636 |
+
"eval_runtime": 7.1789,
|
| 637 |
+
"eval_samples_per_second": 7.243,
|
| 638 |
+
"eval_steps_per_second": 0.975,
|
| 639 |
+
"step": 420
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.5103857566765578,
|
| 643 |
+
"grad_norm": 0.04685925852904185,
|
| 644 |
+
"learning_rate": 0.00011385794572382357,
|
| 645 |
+
"loss": 0.7571,
|
| 646 |
+
"step": 430
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.5103857566765578,
|
| 650 |
+
"eval_loss": 0.660754919052124,
|
| 651 |
+
"eval_runtime": 7.1679,
|
| 652 |
+
"eval_samples_per_second": 7.255,
|
| 653 |
+
"eval_steps_per_second": 0.977,
|
| 654 |
+
"step": 430
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.5222551928783383,
|
| 658 |
+
"grad_norm": 0.05188129116000765,
|
| 659 |
+
"learning_rate": 0.00010973717959097327,
|
| 660 |
+
"loss": 0.7355,
|
| 661 |
+
"step": 440
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.5222551928783383,
|
| 665 |
+
"eval_loss": 0.6598291993141174,
|
| 666 |
+
"eval_runtime": 7.1682,
|
| 667 |
+
"eval_samples_per_second": 7.254,
|
| 668 |
+
"eval_steps_per_second": 0.977,
|
| 669 |
+
"step": 440
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.5341246290801187,
|
| 673 |
+
"grad_norm": 0.04549793036088958,
|
| 674 |
+
"learning_rate": 0.00010559964555200889,
|
| 675 |
+
"loss": 0.7379,
|
| 676 |
+
"step": 450
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.5341246290801187,
|
| 680 |
+
"eval_loss": 0.658793568611145,
|
| 681 |
+
"eval_runtime": 7.1553,
|
| 682 |
+
"eval_samples_per_second": 7.267,
|
| 683 |
+
"eval_steps_per_second": 0.978,
|
| 684 |
+
"step": 450
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.5459940652818991,
|
| 688 |
+
"grad_norm": 0.05162604134236413,
|
| 689 |
+
"learning_rate": 0.0001014524686457223,
|
| 690 |
+
"loss": 0.7363,
|
| 691 |
+
"step": 460
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.5459940652818991,
|
| 695 |
+
"eval_loss": 0.6571336984634399,
|
| 696 |
+
"eval_runtime": 7.156,
|
| 697 |
+
"eval_samples_per_second": 7.267,
|
| 698 |
+
"eval_steps_per_second": 0.978,
|
| 699 |
+
"step": 460
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.5578635014836796,
|
| 703 |
+
"grad_norm": 0.04783457254038487,
|
| 704 |
+
"learning_rate": 9.730279051640112e-05,
|
| 705 |
+
"loss": 0.7177,
|
| 706 |
+
"step": 470
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.5578635014836796,
|
| 710 |
+
"eval_loss": 0.655822217464447,
|
| 711 |
+
"eval_runtime": 7.1579,
|
| 712 |
+
"eval_samples_per_second": 7.265,
|
| 713 |
+
"eval_steps_per_second": 0.978,
|
| 714 |
+
"step": 470
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.56973293768546,
|
| 718 |
+
"grad_norm": 0.04436373760940458,
|
| 719 |
+
"learning_rate": 9.315775711556298e-05,
|
| 720 |
+
"loss": 0.7334,
|
| 721 |
+
"step": 480
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.56973293768546,
|
| 725 |
+
"eval_loss": 0.6540331244468689,
|
| 726 |
+
"eval_runtime": 7.1522,
|
| 727 |
+
"eval_samples_per_second": 7.271,
|
| 728 |
+
"eval_steps_per_second": 0.979,
|
| 729 |
+
"step": 480
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.5816023738872403,
|
| 733 |
+
"grad_norm": 0.04519231140145815,
|
| 734 |
+
"learning_rate": 8.902450639627288e-05,
|
| 735 |
+
"loss": 0.737,
|
| 736 |
+
"step": 490
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.5816023738872403,
|
| 740 |
+
"eval_loss": 0.6534115672111511,
|
| 741 |
+
"eval_runtime": 7.156,
|
| 742 |
+
"eval_samples_per_second": 7.267,
|
| 743 |
+
"eval_steps_per_second": 0.978,
|
| 744 |
+
"step": 490
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5934718100890207,
|
| 748 |
+
"grad_norm": 0.04247565082967707,
|
| 749 |
+
"learning_rate": 8.491015602123368e-05,
|
| 750 |
+
"loss": 0.7413,
|
| 751 |
+
"step": 500
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5934718100890207,
|
| 755 |
+
"eval_loss": 0.6529149413108826,
|
| 756 |
+
"eval_runtime": 7.1578,
|
| 757 |
+
"eval_samples_per_second": 7.265,
|
| 758 |
+
"eval_steps_per_second": 0.978,
|
| 759 |
+
"step": 500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.6053412462908012,
|
| 763 |
+
"grad_norm": 0.048070529053236284,
|
| 764 |
+
"learning_rate": 8.082179110581838e-05,
|
| 765 |
+
"loss": 0.7141,
|
| 766 |
+
"step": 510
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.6053412462908012,
|
| 770 |
+
"eval_loss": 0.651964545249939,
|
| 771 |
+
"eval_runtime": 7.1573,
|
| 772 |
+
"eval_samples_per_second": 7.265,
|
| 773 |
+
"eval_steps_per_second": 0.978,
|
| 774 |
+
"step": 510
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.6172106824925816,
|
| 778 |
+
"grad_norm": 0.04071752578010737,
|
| 779 |
+
"learning_rate": 7.676645201714983e-05,
|
| 780 |
+
"loss": 0.7248,
|
| 781 |
+
"step": 520
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.6172106824925816,
|
| 785 |
+
"eval_loss": 0.6511014103889465,
|
| 786 |
+
"eval_runtime": 7.1566,
|
| 787 |
+
"eval_samples_per_second": 7.266,
|
| 788 |
+
"eval_steps_per_second": 0.978,
|
| 789 |
+
"step": 520
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.629080118694362,
|
| 793 |
+
"grad_norm": 0.04168337358828772,
|
| 794 |
+
"learning_rate": 7.27511222502395e-05,
|
| 795 |
+
"loss": 0.725,
|
| 796 |
+
"step": 530
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.629080118694362,
|
| 800 |
+
"eval_loss": 0.6503429412841797,
|
| 801 |
+
"eval_runtime": 7.1538,
|
| 802 |
+
"eval_samples_per_second": 7.269,
|
| 803 |
+
"eval_steps_per_second": 0.979,
|
| 804 |
+
"step": 530
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.6409495548961425,
|
| 808 |
+
"grad_norm": 0.05260277652628936,
|
| 809 |
+
"learning_rate": 6.878271640206281e-05,
|
| 810 |
+
"loss": 0.7299,
|
| 811 |
+
"step": 540
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.6409495548961425,
|
| 815 |
+
"eval_loss": 0.6499541401863098,
|
| 816 |
+
"eval_runtime": 7.1571,
|
| 817 |
+
"eval_samples_per_second": 7.265,
|
| 818 |
+
"eval_steps_per_second": 0.978,
|
| 819 |
+
"step": 540
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.6528189910979229,
|
| 823 |
+
"grad_norm": 0.04558705692362712,
|
| 824 |
+
"learning_rate": 6.486806826428016e-05,
|
| 825 |
+
"loss": 0.7381,
|
| 826 |
+
"step": 550
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.6528189910979229,
|
| 830 |
+
"eval_loss": 0.6490299701690674,
|
| 831 |
+
"eval_runtime": 7.1586,
|
| 832 |
+
"eval_samples_per_second": 7.264,
|
| 833 |
+
"eval_steps_per_second": 0.978,
|
| 834 |
+
"step": 550
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.6646884272997032,
|
| 838 |
+
"grad_norm": 0.046779386203552986,
|
| 839 |
+
"learning_rate": 6.101391905510889e-05,
|
| 840 |
+
"loss": 0.7052,
|
| 841 |
+
"step": 560
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.6646884272997032,
|
| 845 |
+
"eval_loss": 0.6482182741165161,
|
| 846 |
+
"eval_runtime": 7.1587,
|
| 847 |
+
"eval_samples_per_second": 7.264,
|
| 848 |
+
"eval_steps_per_second": 0.978,
|
| 849 |
+
"step": 560
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6765578635014837,
|
| 853 |
+
"grad_norm": 0.04433666961158006,
|
| 854 |
+
"learning_rate": 5.72269058106111e-05,
|
| 855 |
+
"loss": 0.7267,
|
| 856 |
+
"step": 570
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6765578635014837,
|
| 860 |
+
"eval_loss": 0.6474419832229614,
|
| 861 |
+
"eval_runtime": 7.1585,
|
| 862 |
+
"eval_samples_per_second": 7.264,
|
| 863 |
+
"eval_steps_per_second": 0.978,
|
| 864 |
+
"step": 570
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.6884272997032641,
|
| 868 |
+
"grad_norm": 0.040969171268037584,
|
| 869 |
+
"learning_rate": 5.351354995538859e-05,
|
| 870 |
+
"loss": 0.7217,
|
| 871 |
+
"step": 580
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.6884272997032641,
|
| 875 |
+
"eval_loss": 0.6467857956886292,
|
| 876 |
+
"eval_runtime": 7.1605,
|
| 877 |
+
"eval_samples_per_second": 7.262,
|
| 878 |
+
"eval_steps_per_second": 0.978,
|
| 879 |
+
"step": 580
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.7002967359050445,
|
| 883 |
+
"grad_norm": 0.04326822029385217,
|
| 884 |
+
"learning_rate": 4.988024607236619e-05,
|
| 885 |
+
"loss": 0.7225,
|
| 886 |
+
"step": 590
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.7002967359050445,
|
| 890 |
+
"eval_loss": 0.6460662484169006,
|
| 891 |
+
"eval_runtime": 7.1561,
|
| 892 |
+
"eval_samples_per_second": 7.267,
|
| 893 |
+
"eval_steps_per_second": 0.978,
|
| 894 |
+
"step": 590
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.712166172106825,
|
| 898 |
+
"grad_norm": 0.04151648905301352,
|
| 899 |
+
"learning_rate": 4.633325089100289e-05,
|
| 900 |
+
"loss": 0.7305,
|
| 901 |
+
"step": 600
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.712166172106825,
|
| 905 |
+
"eval_loss": 0.6456657648086548,
|
| 906 |
+
"eval_runtime": 7.1693,
|
| 907 |
+
"eval_samples_per_second": 7.253,
|
| 908 |
+
"eval_steps_per_second": 0.976,
|
| 909 |
+
"step": 600
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.7240356083086054,
|
| 913 |
+
"grad_norm": 0.043211271741518106,
|
| 914 |
+
"learning_rate": 4.287867251289348e-05,
|
| 915 |
+
"loss": 0.7202,
|
| 916 |
+
"step": 610
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.7240356083086054,
|
| 920 |
+
"eval_loss": 0.6450461745262146,
|
| 921 |
+
"eval_runtime": 7.1607,
|
| 922 |
+
"eval_samples_per_second": 7.262,
|
| 923 |
+
"eval_steps_per_second": 0.978,
|
| 924 |
+
"step": 610
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 0.7359050445103857,
|
| 928 |
+
"grad_norm": 0.043394130216544674,
|
| 929 |
+
"learning_rate": 3.952245989331466e-05,
|
| 930 |
+
"loss": 0.7182,
|
| 931 |
+
"step": 620
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.7359050445103857,
|
| 935 |
+
"eval_loss": 0.6443552374839783,
|
| 936 |
+
"eval_runtime": 7.1601,
|
| 937 |
+
"eval_samples_per_second": 7.262,
|
| 938 |
+
"eval_steps_per_second": 0.978,
|
| 939 |
+
"step": 620
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.7477744807121661,
|
| 943 |
+
"grad_norm": 0.0440813623198057,
|
| 944 |
+
"learning_rate": 3.627039259682899e-05,
|
| 945 |
+
"loss": 0.7269,
|
| 946 |
+
"step": 630
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.7477744807121661,
|
| 950 |
+
"eval_loss": 0.6440042853355408,
|
| 951 |
+
"eval_runtime": 7.1528,
|
| 952 |
+
"eval_samples_per_second": 7.27,
|
| 953 |
+
"eval_steps_per_second": 0.979,
|
| 954 |
+
"step": 630
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7596439169139466,
|
| 958 |
+
"grad_norm": 0.04196493736245454,
|
| 959 |
+
"learning_rate": 3.312807084458831e-05,
|
| 960 |
+
"loss": 0.7321,
|
| 961 |
+
"step": 640
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7596439169139466,
|
| 965 |
+
"eval_loss": 0.6436744332313538,
|
| 966 |
+
"eval_runtime": 7.157,
|
| 967 |
+
"eval_samples_per_second": 7.266,
|
| 968 |
+
"eval_steps_per_second": 0.978,
|
| 969 |
+
"step": 640
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.771513353115727,
|
| 973 |
+
"grad_norm": 0.04409502511550107,
|
| 974 |
+
"learning_rate": 3.0100905870475006e-05,
|
| 975 |
+
"loss": 0.7127,
|
| 976 |
+
"step": 650
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.771513353115727,
|
| 980 |
+
"eval_loss": 0.6433111429214478,
|
| 981 |
+
"eval_runtime": 7.1488,
|
| 982 |
+
"eval_samples_per_second": 7.274,
|
| 983 |
+
"eval_steps_per_second": 0.979,
|
| 984 |
+
"step": 650
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.7833827893175074,
|
| 988 |
+
"grad_norm": 0.04613176858045456,
|
| 989 |
+
"learning_rate": 2.7194110602689026e-05,
|
| 990 |
+
"loss": 0.7084,
|
| 991 |
+
"step": 660
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.7833827893175074,
|
| 995 |
+
"eval_loss": 0.6427409648895264,
|
| 996 |
+
"eval_runtime": 7.1651,
|
| 997 |
+
"eval_samples_per_second": 7.257,
|
| 998 |
+
"eval_steps_per_second": 0.977,
|
| 999 |
+
"step": 660
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.7952522255192879,
|
| 1003 |
+
"grad_norm": 0.04107118214802092,
|
| 1004 |
+
"learning_rate": 2.4412690686827e-05,
|
| 1005 |
+
"loss": 0.7318,
|
| 1006 |
+
"step": 670
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.7952522255192879,
|
| 1010 |
+
"eval_loss": 0.6424255967140198,
|
| 1011 |
+
"eval_runtime": 7.151,
|
| 1012 |
+
"eval_samples_per_second": 7.272,
|
| 1013 |
+
"eval_steps_per_second": 0.979,
|
| 1014 |
+
"step": 670
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.8071216617210683,
|
| 1018 |
+
"grad_norm": 0.04076823498258544,
|
| 1019 |
+
"learning_rate": 2.1761435865912296e-05,
|
| 1020 |
+
"loss": 0.7111,
|
| 1021 |
+
"step": 680
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.8071216617210683,
|
| 1025 |
+
"eval_loss": 0.6420266032218933,
|
| 1026 |
+
"eval_runtime": 7.1562,
|
| 1027 |
+
"eval_samples_per_second": 7.266,
|
| 1028 |
+
"eval_steps_per_second": 0.978,
|
| 1029 |
+
"step": 680
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.8189910979228486,
|
| 1033 |
+
"grad_norm": 0.0402931102905212,
|
| 1034 |
+
"learning_rate": 1.9244911732219918e-05,
|
| 1035 |
+
"loss": 0.7173,
|
| 1036 |
+
"step": 690
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.8189910979228486,
|
| 1040 |
+
"eval_loss": 0.6417113542556763,
|
| 1041 |
+
"eval_runtime": 7.1559,
|
| 1042 |
+
"eval_samples_per_second": 7.267,
|
| 1043 |
+
"eval_steps_per_second": 0.978,
|
| 1044 |
+
"step": 690
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.8308605341246291,
|
| 1048 |
+
"grad_norm": 0.040268557762468446,
|
| 1049 |
+
"learning_rate": 1.6867451865100414e-05,
|
| 1050 |
+
"loss": 0.7187,
|
| 1051 |
+
"step": 700
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.8308605341246291,
|
| 1055 |
+
"eval_loss": 0.6414252519607544,
|
| 1056 |
+
"eval_runtime": 7.1614,
|
| 1057 |
+
"eval_samples_per_second": 7.261,
|
| 1058 |
+
"eval_steps_per_second": 0.977,
|
| 1059 |
+
"step": 700
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.8427299703264095,
|
| 1063 |
+
"grad_norm": 0.03910974935669944,
|
| 1064 |
+
"learning_rate": 1.4633150368341153e-05,
|
| 1065 |
+
"loss": 0.7003,
|
| 1066 |
+
"step": 710
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.8427299703264095,
|
| 1070 |
+
"eval_loss": 0.6411222815513611,
|
| 1071 |
+
"eval_runtime": 7.158,
|
| 1072 |
+
"eval_samples_per_second": 7.265,
|
| 1073 |
+
"eval_steps_per_second": 0.978,
|
| 1074 |
+
"step": 710
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.8545994065281899,
|
| 1078 |
+
"grad_norm": 0.041327420976484716,
|
| 1079 |
+
"learning_rate": 1.2545854819916646e-05,
|
| 1080 |
+
"loss": 0.7146,
|
| 1081 |
+
"step": 720
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.8545994065281899,
|
| 1085 |
+
"eval_loss": 0.6409789323806763,
|
| 1086 |
+
"eval_runtime": 7.1632,
|
| 1087 |
+
"eval_samples_per_second": 7.259,
|
| 1088 |
+
"eval_steps_per_second": 0.977,
|
| 1089 |
+
"step": 720
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.8664688427299704,
|
| 1093 |
+
"grad_norm": 0.0415664844879509,
|
| 1094 |
+
"learning_rate": 1.0609159646268506e-05,
|
| 1095 |
+
"loss": 0.7326,
|
| 1096 |
+
"step": 730
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.8664688427299704,
|
| 1100 |
+
"eval_loss": 0.640900194644928,
|
| 1101 |
+
"eval_runtime": 7.1625,
|
| 1102 |
+
"eval_samples_per_second": 7.26,
|
| 1103 |
+
"eval_steps_per_second": 0.977,
|
| 1104 |
+
"step": 730
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.8783382789317508,
|
| 1108 |
+
"grad_norm": 0.040948908512356046,
|
| 1109 |
+
"learning_rate": 8.82639993252482e-06,
|
| 1110 |
+
"loss": 0.735,
|
| 1111 |
+
"step": 740
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 0.8783382789317508,
|
| 1115 |
+
"eval_loss": 0.6406245827674866,
|
| 1116 |
+
"eval_runtime": 7.1641,
|
| 1117 |
+
"eval_samples_per_second": 7.258,
|
| 1118 |
+
"eval_steps_per_second": 0.977,
|
| 1119 |
+
"step": 740
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.8902077151335311,
|
| 1123 |
+
"grad_norm": 0.039353242405877606,
|
| 1124 |
+
"learning_rate": 7.20064567931813e-06,
|
| 1125 |
+
"loss": 0.7072,
|
| 1126 |
+
"step": 750
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 0.8902077151335311,
|
| 1130 |
+
"eval_loss": 0.6405530571937561,
|
| 1131 |
+
"eval_runtime": 7.1654,
|
| 1132 |
+
"eval_samples_per_second": 7.257,
|
| 1133 |
+
"eval_steps_per_second": 0.977,
|
| 1134 |
+
"step": 750
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 0.9020771513353115,
|
| 1138 |
+
"grad_norm": 0.040045103519338765,
|
| 1139 |
+
"learning_rate": 5.734696516092253e-06,
|
| 1140 |
+
"loss": 0.7242,
|
| 1141 |
+
"step": 760
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 0.9020771513353115,
|
| 1145 |
+
"eval_loss": 0.6403719186782837,
|
| 1146 |
+
"eval_runtime": 7.1641,
|
| 1147 |
+
"eval_samples_per_second": 7.258,
|
| 1148 |
+
"eval_steps_per_second": 0.977,
|
| 1149 |
+
"step": 760
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.913946587537092,
|
| 1153 |
+
"grad_norm": 0.03936062921235335,
|
| 1154 |
+
"learning_rate": 4.431076880001439e-06,
|
| 1155 |
+
"loss": 0.725,
|
| 1156 |
+
"step": 770
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.913946587537092,
|
| 1160 |
+
"eval_loss": 0.6402689218521118,
|
| 1161 |
+
"eval_runtime": 7.1586,
|
| 1162 |
+
"eval_samples_per_second": 7.264,
|
| 1163 |
+
"eval_steps_per_second": 0.978,
|
| 1164 |
+
"step": 770
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.9258160237388724,
|
| 1168 |
+
"grad_norm": 0.038795606374443216,
|
| 1169 |
+
"learning_rate": 3.292031668704398e-06,
|
| 1170 |
+
"loss": 0.7051,
|
| 1171 |
+
"step": 780
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.9258160237388724,
|
| 1175 |
+
"eval_loss": 0.6402180790901184,
|
| 1176 |
+
"eval_runtime": 7.1753,
|
| 1177 |
+
"eval_samples_per_second": 7.247,
|
| 1178 |
+
"eval_steps_per_second": 0.976,
|
| 1179 |
+
"step": 780
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.9376854599406528,
|
| 1183 |
+
"grad_norm": 0.04044906534894689,
|
| 1184 |
+
"learning_rate": 2.3195223745392737e-06,
|
| 1185 |
+
"loss": 0.7187,
|
| 1186 |
+
"step": 790
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.9376854599406528,
|
| 1190 |
+
"eval_loss": 0.6401559114456177,
|
| 1191 |
+
"eval_runtime": 7.178,
|
| 1192 |
+
"eval_samples_per_second": 7.244,
|
| 1193 |
+
"eval_steps_per_second": 0.975,
|
| 1194 |
+
"step": 790
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.9495548961424333,
|
| 1198 |
+
"grad_norm": 0.03933803718026143,
|
| 1199 |
+
"learning_rate": 1.5152237067365239e-06,
|
| 1200 |
+
"loss": 0.7185,
|
| 1201 |
+
"step": 800
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.9495548961424333,
|
| 1205 |
+
"eval_loss": 0.6400689482688904,
|
| 1206 |
+
"eval_runtime": 7.1614,
|
| 1207 |
+
"eval_samples_per_second": 7.261,
|
| 1208 |
+
"eval_steps_per_second": 0.977,
|
| 1209 |
+
"step": 800
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.9614243323442137,
|
| 1213 |
+
"grad_norm": 0.03894023495076019,
|
| 1214 |
+
"learning_rate": 8.805207074865873e-07,
|
| 1215 |
+
"loss": 0.7213,
|
| 1216 |
+
"step": 810
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.9614243323442137,
|
| 1220 |
+
"eval_loss": 0.6400617957115173,
|
| 1221 |
+
"eval_runtime": 7.1732,
|
| 1222 |
+
"eval_samples_per_second": 7.249,
|
| 1223 |
+
"eval_steps_per_second": 0.976,
|
| 1224 |
+
"step": 810
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.973293768545994,
|
| 1228 |
+
"grad_norm": 0.03930788498385229,
|
| 1229 |
+
"learning_rate": 4.165063668285396e-07,
|
| 1230 |
+
"loss": 0.7192,
|
| 1231 |
+
"step": 820
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.973293768545994,
|
| 1235 |
+
"eval_loss": 0.6400607228279114,
|
| 1236 |
+
"eval_runtime": 7.1626,
|
| 1237 |
+
"eval_samples_per_second": 7.26,
|
| 1238 |
+
"eval_steps_per_second": 0.977,
|
| 1239 |
+
"step": 820
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 0.9851632047477745,
|
| 1243 |
+
"grad_norm": 0.03939532170183159,
|
| 1244 |
+
"learning_rate": 1.2397974046707283e-07,
|
| 1245 |
+
"loss": 0.7214,
|
| 1246 |
+
"step": 830
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.9851632047477745,
|
| 1250 |
+
"eval_loss": 0.6400620341300964,
|
| 1251 |
+
"eval_runtime": 7.1644,
|
| 1252 |
+
"eval_samples_per_second": 7.258,
|
| 1253 |
+
"eval_steps_per_second": 0.977,
|
| 1254 |
+
"step": 830
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.9970326409495549,
|
| 1258 |
+
"grad_norm": 0.04057041649987638,
|
| 1259 |
+
"learning_rate": 3.444573758937253e-09,
|
| 1260 |
+
"loss": 0.7317,
|
| 1261 |
+
"step": 840
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.9970326409495549,
|
| 1265 |
+
"eval_loss": 0.640075147151947,
|
| 1266 |
+
"eval_runtime": 7.1623,
|
| 1267 |
+
"eval_samples_per_second": 7.26,
|
| 1268 |
+
"eval_steps_per_second": 0.977,
|
| 1269 |
+
"step": 840
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.9994065281899109,
|
| 1273 |
+
"step": 842,
|
| 1274 |
+
"total_flos": 6.615025481416704e+16,
|
| 1275 |
+
"train_loss": 0.6960575034103031,
|
| 1276 |
+
"train_runtime": 22091.905,
|
| 1277 |
+
"train_samples_per_second": 2.44,
|
| 1278 |
+
"train_steps_per_second": 0.038
|
| 1279 |
+
}
|
| 1280 |
+
],
|
| 1281 |
+
"logging_steps": 10,
|
| 1282 |
+
"max_steps": 842,
|
| 1283 |
+
"num_input_tokens_seen": 0,
|
| 1284 |
+
"num_train_epochs": 1,
|
| 1285 |
+
"save_steps": 10,
|
| 1286 |
+
"stateful_callbacks": {
|
| 1287 |
+
"TrainerControl": {
|
| 1288 |
+
"args": {
|
| 1289 |
+
"should_epoch_stop": false,
|
| 1290 |
+
"should_evaluate": false,
|
| 1291 |
+
"should_log": false,
|
| 1292 |
+
"should_save": true,
|
| 1293 |
+
"should_training_stop": true
|
| 1294 |
+
},
|
| 1295 |
+
"attributes": {}
|
| 1296 |
+
}
|
| 1297 |
+
},
|
| 1298 |
+
"total_flos": 6.615025481416704e+16,
|
| 1299 |
+
"train_batch_size": 8,
|
| 1300 |
+
"trial_name": null,
|
| 1301 |
+
"trial_params": null
|
| 1302 |
+
}
|