Model save
Browse files- README.md +58 -0
- all_results.json +9 -0
- config.json +1 -1
- generation_config.json +14 -0
- train_results.json +9 -0
- trainer_state.json +2002 -0
README.md
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen2.5-1.5B-Instruct
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Qwen2.5-Argunaut-1-1.5B-SFT-dev0
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Qwen2.5-Argunaut-1-1.5B-SFT-dev0
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct).
|
| 15 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
+
|
| 17 |
+
## Quick start
|
| 18 |
+
|
| 19 |
+
```python
|
| 20 |
+
from transformers import pipeline
|
| 21 |
+
|
| 22 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 23 |
+
generator = pipeline("text-generation", model="DebateLabKIT/Qwen2.5-Argunaut-1-1.5B-SFT-dev0", 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/ggbetz/argunauts-training/runs/sai3aev5)
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with SFT.
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.14.0
|
| 38 |
+
- Transformers: 4.46.3
|
| 39 |
+
- Pytorch: 2.4.1
|
| 40 |
+
- Datasets: 3.1.0
|
| 41 |
+
- Tokenizers: 0.20.3
|
| 42 |
+
|
| 43 |
+
## Citations
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
Cite TRL as:
|
| 48 |
+
|
| 49 |
+
```bibtex
|
| 50 |
+
@misc{vonwerra2022trl,
|
| 51 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 52 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
|
| 53 |
+
year = 2020,
|
| 54 |
+
journal = {GitHub repository},
|
| 55 |
+
publisher = {GitHub},
|
| 56 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 57 |
+
}
|
| 58 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1267133463920640.0,
|
| 4 |
+
"train_loss": 0.6670485469077131,
|
| 5 |
+
"train_runtime": 18244.3099,
|
| 6 |
+
"train_samples": 1000000,
|
| 7 |
+
"train_samples_per_second": 9.746,
|
| 8 |
+
"train_steps_per_second": 0.076
|
| 9 |
+
}
|
config.json
CHANGED
|
@@ -23,7 +23,7 @@
|
|
| 23 |
"tie_word_embeddings": true,
|
| 24 |
"torch_dtype": "bfloat16",
|
| 25 |
"transformers_version": "4.46.3",
|
| 26 |
-
"use_cache":
|
| 27 |
"use_sliding_window": false,
|
| 28 |
"vocab_size": 151936
|
| 29 |
}
|
|
|
|
| 23 |
"tie_word_embeddings": true,
|
| 24 |
"torch_dtype": "bfloat16",
|
| 25 |
"transformers_version": "4.46.3",
|
| 26 |
+
"use_cache": true,
|
| 27 |
"use_sliding_window": false,
|
| 28 |
"vocab_size": 151936
|
| 29 |
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.46.3"
|
| 14 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1267133463920640.0,
|
| 4 |
+
"train_loss": 0.6670485469077131,
|
| 5 |
+
"train_runtime": 18244.3099,
|
| 6 |
+
"train_samples": 1000000,
|
| 7 |
+
"train_samples_per_second": 9.746,
|
| 8 |
+
"train_steps_per_second": 0.076
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2002 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1390,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0035971223021582736,
|
| 13 |
+
"grad_norm": 5.020955964884568,
|
| 14 |
+
"learning_rate": 1.7985611510791368e-07,
|
| 15 |
+
"loss": 1.1944,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007194244604316547,
|
| 20 |
+
"grad_norm": 4.938024747588621,
|
| 21 |
+
"learning_rate": 3.5971223021582736e-07,
|
| 22 |
+
"loss": 1.213,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01079136690647482,
|
| 27 |
+
"grad_norm": 4.729140166656345,
|
| 28 |
+
"learning_rate": 5.39568345323741e-07,
|
| 29 |
+
"loss": 1.2154,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014388489208633094,
|
| 34 |
+
"grad_norm": 4.486827019767407,
|
| 35 |
+
"learning_rate": 7.194244604316547e-07,
|
| 36 |
+
"loss": 1.1966,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.017985611510791366,
|
| 41 |
+
"grad_norm": 2.8354701196618293,
|
| 42 |
+
"learning_rate": 8.992805755395684e-07,
|
| 43 |
+
"loss": 1.1617,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02158273381294964,
|
| 48 |
+
"grad_norm": 1.502509909859451,
|
| 49 |
+
"learning_rate": 1.079136690647482e-06,
|
| 50 |
+
"loss": 1.1554,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.025179856115107913,
|
| 55 |
+
"grad_norm": 0.8955286838566213,
|
| 56 |
+
"learning_rate": 1.2589928057553958e-06,
|
| 57 |
+
"loss": 1.103,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02877697841726619,
|
| 62 |
+
"grad_norm": 0.7456236862794698,
|
| 63 |
+
"learning_rate": 1.4388489208633094e-06,
|
| 64 |
+
"loss": 1.1106,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03237410071942446,
|
| 69 |
+
"grad_norm": 0.7017287601950278,
|
| 70 |
+
"learning_rate": 1.618705035971223e-06,
|
| 71 |
+
"loss": 1.0751,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03597122302158273,
|
| 76 |
+
"grad_norm": 0.5721443296844856,
|
| 77 |
+
"learning_rate": 1.7985611510791368e-06,
|
| 78 |
+
"loss": 1.0696,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.039568345323741004,
|
| 83 |
+
"grad_norm": 0.4464262507331005,
|
| 84 |
+
"learning_rate": 1.9784172661870504e-06,
|
| 85 |
+
"loss": 1.0304,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04316546762589928,
|
| 90 |
+
"grad_norm": 0.4092857312926078,
|
| 91 |
+
"learning_rate": 2.158273381294964e-06,
|
| 92 |
+
"loss": 1.0121,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.046762589928057555,
|
| 97 |
+
"grad_norm": 0.3578204568994334,
|
| 98 |
+
"learning_rate": 2.3381294964028776e-06,
|
| 99 |
+
"loss": 0.9918,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.050359712230215826,
|
| 104 |
+
"grad_norm": 0.343624490661988,
|
| 105 |
+
"learning_rate": 2.5179856115107916e-06,
|
| 106 |
+
"loss": 0.9676,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.0539568345323741,
|
| 111 |
+
"grad_norm": 0.3052531774113653,
|
| 112 |
+
"learning_rate": 2.6978417266187052e-06,
|
| 113 |
+
"loss": 0.9529,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05755395683453238,
|
| 118 |
+
"grad_norm": 0.29429542067049574,
|
| 119 |
+
"learning_rate": 2.877697841726619e-06,
|
| 120 |
+
"loss": 0.9197,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06115107913669065,
|
| 125 |
+
"grad_norm": 0.27942488562810386,
|
| 126 |
+
"learning_rate": 3.0575539568345324e-06,
|
| 127 |
+
"loss": 0.8649,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06474820143884892,
|
| 132 |
+
"grad_norm": 0.2754960077654889,
|
| 133 |
+
"learning_rate": 3.237410071942446e-06,
|
| 134 |
+
"loss": 0.8803,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0683453237410072,
|
| 139 |
+
"grad_norm": 0.2637478480095364,
|
| 140 |
+
"learning_rate": 3.4172661870503596e-06,
|
| 141 |
+
"loss": 0.833,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07194244604316546,
|
| 146 |
+
"grad_norm": 0.2588869176654825,
|
| 147 |
+
"learning_rate": 3.5971223021582737e-06,
|
| 148 |
+
"loss": 0.824,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07553956834532374,
|
| 153 |
+
"grad_norm": 0.25809976308965515,
|
| 154 |
+
"learning_rate": 3.7769784172661873e-06,
|
| 155 |
+
"loss": 0.8134,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07913669064748201,
|
| 160 |
+
"grad_norm": 0.2280622185227855,
|
| 161 |
+
"learning_rate": 3.956834532374101e-06,
|
| 162 |
+
"loss": 0.8173,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08273381294964029,
|
| 167 |
+
"grad_norm": 0.23779953523720807,
|
| 168 |
+
"learning_rate": 4.1366906474820145e-06,
|
| 169 |
+
"loss": 0.7979,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08633093525179857,
|
| 174 |
+
"grad_norm": 0.22506106711683146,
|
| 175 |
+
"learning_rate": 4.316546762589928e-06,
|
| 176 |
+
"loss": 0.7681,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08992805755395683,
|
| 181 |
+
"grad_norm": 0.2175874711615906,
|
| 182 |
+
"learning_rate": 4.496402877697842e-06,
|
| 183 |
+
"loss": 0.7916,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09352517985611511,
|
| 188 |
+
"grad_norm": 0.2118293459237304,
|
| 189 |
+
"learning_rate": 4.676258992805755e-06,
|
| 190 |
+
"loss": 0.7798,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.09712230215827339,
|
| 195 |
+
"grad_norm": 0.22010121805042218,
|
| 196 |
+
"learning_rate": 4.856115107913669e-06,
|
| 197 |
+
"loss": 0.7218,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10071942446043165,
|
| 202 |
+
"grad_norm": 0.20337355284622044,
|
| 203 |
+
"learning_rate": 4.999992116938572e-06,
|
| 204 |
+
"loss": 0.7329,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10431654676258993,
|
| 209 |
+
"grad_norm": 0.19294611128703623,
|
| 210 |
+
"learning_rate": 4.999716215008542e-06,
|
| 211 |
+
"loss": 0.7343,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1079136690647482,
|
| 216 |
+
"grad_norm": 0.18569152586574716,
|
| 217 |
+
"learning_rate": 4.999046209719832e-06,
|
| 218 |
+
"loss": 0.7078,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11151079136690648,
|
| 223 |
+
"grad_norm": 0.19184024245666095,
|
| 224 |
+
"learning_rate": 4.997982206704965e-06,
|
| 225 |
+
"loss": 0.7279,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11510791366906475,
|
| 230 |
+
"grad_norm": 0.1943879874060985,
|
| 231 |
+
"learning_rate": 4.996524373713848e-06,
|
| 232 |
+
"loss": 0.735,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.11870503597122302,
|
| 237 |
+
"grad_norm": 0.188675637921693,
|
| 238 |
+
"learning_rate": 4.994672940587324e-06,
|
| 239 |
+
"loss": 0.7113,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.1223021582733813,
|
| 244 |
+
"grad_norm": 0.18883128374108282,
|
| 245 |
+
"learning_rate": 4.992428199220931e-06,
|
| 246 |
+
"loss": 0.718,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12589928057553956,
|
| 251 |
+
"grad_norm": 0.18714736466453813,
|
| 252 |
+
"learning_rate": 4.989790503518888e-06,
|
| 253 |
+
"loss": 0.7153,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.12949640287769784,
|
| 258 |
+
"grad_norm": 0.18374854258433984,
|
| 259 |
+
"learning_rate": 4.986760269338294e-06,
|
| 260 |
+
"loss": 0.7244,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13309352517985612,
|
| 265 |
+
"grad_norm": 0.18743762170823544,
|
| 266 |
+
"learning_rate": 4.983337974423567e-06,
|
| 267 |
+
"loss": 0.7239,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1366906474820144,
|
| 272 |
+
"grad_norm": 0.1849033708626242,
|
| 273 |
+
"learning_rate": 4.979524158331123e-06,
|
| 274 |
+
"loss": 0.7115,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.14028776978417265,
|
| 279 |
+
"grad_norm": 0.18641479765213534,
|
| 280 |
+
"learning_rate": 4.975319422344308e-06,
|
| 281 |
+
"loss": 0.705,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14388489208633093,
|
| 286 |
+
"grad_norm": 0.18977732323761207,
|
| 287 |
+
"learning_rate": 4.970724429378602e-06,
|
| 288 |
+
"loss": 0.7008,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1474820143884892,
|
| 293 |
+
"grad_norm": 0.18827613594876644,
|
| 294 |
+
"learning_rate": 4.9657399038771045e-06,
|
| 295 |
+
"loss": 0.6916,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1510791366906475,
|
| 300 |
+
"grad_norm": 0.1770962322034755,
|
| 301 |
+
"learning_rate": 4.960366631696317e-06,
|
| 302 |
+
"loss": 0.6861,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15467625899280577,
|
| 307 |
+
"grad_norm": 0.18186648674203168,
|
| 308 |
+
"learning_rate": 4.954605459982248e-06,
|
| 309 |
+
"loss": 0.6758,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.15827338129496402,
|
| 314 |
+
"grad_norm": 0.1766656970929977,
|
| 315 |
+
"learning_rate": 4.9484572970368516e-06,
|
| 316 |
+
"loss": 0.6944,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1618705035971223,
|
| 321 |
+
"grad_norm": 0.18242864024285693,
|
| 322 |
+
"learning_rate": 4.941923112174826e-06,
|
| 323 |
+
"loss": 0.703,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16546762589928057,
|
| 328 |
+
"grad_norm": 0.18130495443681138,
|
| 329 |
+
"learning_rate": 4.935003935570789e-06,
|
| 330 |
+
"loss": 0.6894,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.16906474820143885,
|
| 335 |
+
"grad_norm": 0.17185494495136486,
|
| 336 |
+
"learning_rate": 4.9277008580968665e-06,
|
| 337 |
+
"loss": 0.6786,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17266187050359713,
|
| 342 |
+
"grad_norm": 0.17745476735656734,
|
| 343 |
+
"learning_rate": 4.920015031150702e-06,
|
| 344 |
+
"loss": 0.6885,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.17625899280575538,
|
| 349 |
+
"grad_norm": 0.17746918416706206,
|
| 350 |
+
"learning_rate": 4.911947666473932e-06,
|
| 351 |
+
"loss": 0.6566,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.17985611510791366,
|
| 356 |
+
"grad_norm": 0.17890622279080073,
|
| 357 |
+
"learning_rate": 4.903500035961139e-06,
|
| 358 |
+
"loss": 0.6607,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18345323741007194,
|
| 363 |
+
"grad_norm": 0.17911194474517148,
|
| 364 |
+
"learning_rate": 4.894673471459331e-06,
|
| 365 |
+
"loss": 0.6955,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.18705035971223022,
|
| 370 |
+
"grad_norm": 0.17868419774364908,
|
| 371 |
+
"learning_rate": 4.885469364557956e-06,
|
| 372 |
+
"loss": 0.6888,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1906474820143885,
|
| 377 |
+
"grad_norm": 0.18451377818125733,
|
| 378 |
+
"learning_rate": 4.8758891663695165e-06,
|
| 379 |
+
"loss": 0.6595,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19424460431654678,
|
| 384 |
+
"grad_norm": 0.17946177717246753,
|
| 385 |
+
"learning_rate": 4.865934387300776e-06,
|
| 386 |
+
"loss": 0.684,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.19784172661870503,
|
| 391 |
+
"grad_norm": 0.19613942764348846,
|
| 392 |
+
"learning_rate": 4.8556065968146385e-06,
|
| 393 |
+
"loss": 0.6819,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2014388489208633,
|
| 398 |
+
"grad_norm": 0.17900557510693368,
|
| 399 |
+
"learning_rate": 4.844907423182699e-06,
|
| 400 |
+
"loss": 0.6779,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.20503597122302158,
|
| 405 |
+
"grad_norm": 0.17655099392118984,
|
| 406 |
+
"learning_rate": 4.833838553228547e-06,
|
| 407 |
+
"loss": 0.6598,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.20863309352517986,
|
| 412 |
+
"grad_norm": 0.17383906870606797,
|
| 413 |
+
"learning_rate": 4.822401732061802e-06,
|
| 414 |
+
"loss": 0.6707,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.21223021582733814,
|
| 419 |
+
"grad_norm": 0.18068285893162356,
|
| 420 |
+
"learning_rate": 4.810598762803e-06,
|
| 421 |
+
"loss": 0.6699,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2158273381294964,
|
| 426 |
+
"grad_norm": 0.17183382610494244,
|
| 427 |
+
"learning_rate": 4.798431506299303e-06,
|
| 428 |
+
"loss": 0.6578,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.21942446043165467,
|
| 433 |
+
"grad_norm": 0.18275661567359278,
|
| 434 |
+
"learning_rate": 4.785901880831124e-06,
|
| 435 |
+
"loss": 0.627,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22302158273381295,
|
| 440 |
+
"grad_norm": 0.17659347495851926,
|
| 441 |
+
"learning_rate": 4.773011861809694e-06,
|
| 442 |
+
"loss": 0.6695,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.22661870503597123,
|
| 447 |
+
"grad_norm": 0.17029323062013704,
|
| 448 |
+
"learning_rate": 4.759763481465611e-06,
|
| 449 |
+
"loss": 0.6686,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.2302158273381295,
|
| 454 |
+
"grad_norm": 0.17470042602053984,
|
| 455 |
+
"learning_rate": 4.746158828528457e-06,
|
| 456 |
+
"loss": 0.6387,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.23381294964028776,
|
| 461 |
+
"grad_norm": 0.1717697065536718,
|
| 462 |
+
"learning_rate": 4.73220004789747e-06,
|
| 463 |
+
"loss": 0.6454,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.23741007194244604,
|
| 468 |
+
"grad_norm": 0.17271020357114406,
|
| 469 |
+
"learning_rate": 4.717889340303399e-06,
|
| 470 |
+
"loss": 0.6409,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24100719424460432,
|
| 475 |
+
"grad_norm": 0.17563690998151346,
|
| 476 |
+
"learning_rate": 4.703228961961524e-06,
|
| 477 |
+
"loss": 0.654,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2446043165467626,
|
| 482 |
+
"grad_norm": 0.17294921658141785,
|
| 483 |
+
"learning_rate": 4.6882212242159555e-06,
|
| 484 |
+
"loss": 0.6427,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.24820143884892087,
|
| 489 |
+
"grad_norm": 0.17486817689111733,
|
| 490 |
+
"learning_rate": 4.672868493175219e-06,
|
| 491 |
+
"loss": 0.648,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2517985611510791,
|
| 496 |
+
"grad_norm": 0.17267103530941283,
|
| 497 |
+
"learning_rate": 4.657173189339222e-06,
|
| 498 |
+
"loss": 0.643,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.25539568345323743,
|
| 503 |
+
"grad_norm": 0.17335584700693588,
|
| 504 |
+
"learning_rate": 4.64113778721764e-06,
|
| 505 |
+
"loss": 0.6486,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2589928057553957,
|
| 510 |
+
"grad_norm": 0.16915911050050117,
|
| 511 |
+
"learning_rate": 4.624764814939785e-06,
|
| 512 |
+
"loss": 0.6618,
|
| 513 |
+
"step": 360
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.26258992805755393,
|
| 517 |
+
"grad_norm": 0.1718246313112584,
|
| 518 |
+
"learning_rate": 4.608056853856021e-06,
|
| 519 |
+
"loss": 0.6727,
|
| 520 |
+
"step": 365
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.26618705035971224,
|
| 524 |
+
"grad_norm": 0.17570002762242526,
|
| 525 |
+
"learning_rate": 4.591016538130796e-06,
|
| 526 |
+
"loss": 0.6409,
|
| 527 |
+
"step": 370
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2697841726618705,
|
| 531 |
+
"grad_norm": 0.17190700381660223,
|
| 532 |
+
"learning_rate": 4.573646554327336e-06,
|
| 533 |
+
"loss": 0.6429,
|
| 534 |
+
"step": 375
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2733812949640288,
|
| 538 |
+
"grad_norm": 0.1710399115293668,
|
| 539 |
+
"learning_rate": 4.555949640984087e-06,
|
| 540 |
+
"loss": 0.633,
|
| 541 |
+
"step": 380
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.27697841726618705,
|
| 545 |
+
"grad_norm": 0.17294539558410943,
|
| 546 |
+
"learning_rate": 4.537928588182955e-06,
|
| 547 |
+
"loss": 0.6506,
|
| 548 |
+
"step": 385
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2805755395683453,
|
| 552 |
+
"grad_norm": 0.17037231572109038,
|
| 553 |
+
"learning_rate": 4.519586237109431e-06,
|
| 554 |
+
"loss": 0.6518,
|
| 555 |
+
"step": 390
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2841726618705036,
|
| 559 |
+
"grad_norm": 0.18232442328302445,
|
| 560 |
+
"learning_rate": 4.500925479604645e-06,
|
| 561 |
+
"loss": 0.6303,
|
| 562 |
+
"step": 395
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.28776978417266186,
|
| 566 |
+
"grad_norm": 0.1735924546035651,
|
| 567 |
+
"learning_rate": 4.481949257709442e-06,
|
| 568 |
+
"loss": 0.6471,
|
| 569 |
+
"step": 400
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.29136690647482016,
|
| 573 |
+
"grad_norm": 0.1785525245588453,
|
| 574 |
+
"learning_rate": 4.462660563200545e-06,
|
| 575 |
+
"loss": 0.664,
|
| 576 |
+
"step": 405
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2949640287769784,
|
| 580 |
+
"grad_norm": 0.17152892007141038,
|
| 581 |
+
"learning_rate": 4.44306243711887e-06,
|
| 582 |
+
"loss": 0.651,
|
| 583 |
+
"step": 410
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.29856115107913667,
|
| 587 |
+
"grad_norm": 0.18187119520437112,
|
| 588 |
+
"learning_rate": 4.423157969290081e-06,
|
| 589 |
+
"loss": 0.6772,
|
| 590 |
+
"step": 415
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.302158273381295,
|
| 594 |
+
"grad_norm": 0.17859792680665468,
|
| 595 |
+
"learning_rate": 4.402950297837449e-06,
|
| 596 |
+
"loss": 0.6814,
|
| 597 |
+
"step": 420
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3057553956834532,
|
| 601 |
+
"grad_norm": 0.1663311727852307,
|
| 602 |
+
"learning_rate": 4.382442608687097e-06,
|
| 603 |
+
"loss": 0.6315,
|
| 604 |
+
"step": 425
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.30935251798561153,
|
| 608 |
+
"grad_norm": 0.1701214320556193,
|
| 609 |
+
"learning_rate": 4.361638135065711e-06,
|
| 610 |
+
"loss": 0.6539,
|
| 611 |
+
"step": 430
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3129496402877698,
|
| 615 |
+
"grad_norm": 0.17523250297062218,
|
| 616 |
+
"learning_rate": 4.34054015699079e-06,
|
| 617 |
+
"loss": 0.6667,
|
| 618 |
+
"step": 435
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.31654676258992803,
|
| 622 |
+
"grad_norm": 0.19076203006305673,
|
| 623 |
+
"learning_rate": 4.3191520007535235e-06,
|
| 624 |
+
"loss": 0.6398,
|
| 625 |
+
"step": 440
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.32014388489208634,
|
| 629 |
+
"grad_norm": 0.17133961624985056,
|
| 630 |
+
"learning_rate": 4.297477038394368e-06,
|
| 631 |
+
"loss": 0.6535,
|
| 632 |
+
"step": 445
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3237410071942446,
|
| 636 |
+
"grad_norm": 0.16928107408174564,
|
| 637 |
+
"learning_rate": 4.275518687171418e-06,
|
| 638 |
+
"loss": 0.6531,
|
| 639 |
+
"step": 450
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3273381294964029,
|
| 643 |
+
"grad_norm": 0.16431975083654335,
|
| 644 |
+
"learning_rate": 4.2532804090216374e-06,
|
| 645 |
+
"loss": 0.6542,
|
| 646 |
+
"step": 455
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.33093525179856115,
|
| 650 |
+
"grad_norm": 0.1720650129283096,
|
| 651 |
+
"learning_rate": 4.230765710015058e-06,
|
| 652 |
+
"loss": 0.6582,
|
| 653 |
+
"step": 460
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3345323741007194,
|
| 657 |
+
"grad_norm": 0.16337467117780707,
|
| 658 |
+
"learning_rate": 4.2079781398020155e-06,
|
| 659 |
+
"loss": 0.6221,
|
| 660 |
+
"step": 465
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3381294964028777,
|
| 664 |
+
"grad_norm": 0.17958123058890538,
|
| 665 |
+
"learning_rate": 4.184921291053511e-06,
|
| 666 |
+
"loss": 0.6578,
|
| 667 |
+
"step": 470
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.34172661870503596,
|
| 671 |
+
"grad_norm": 0.17044537787660088,
|
| 672 |
+
"learning_rate": 4.161598798894795e-06,
|
| 673 |
+
"loss": 0.6591,
|
| 674 |
+
"step": 475
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.34532374100719426,
|
| 678 |
+
"grad_norm": 0.162676186545111,
|
| 679 |
+
"learning_rate": 4.1380143403322546e-06,
|
| 680 |
+
"loss": 0.6384,
|
| 681 |
+
"step": 480
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3489208633093525,
|
| 685 |
+
"grad_norm": 0.1694294650638159,
|
| 686 |
+
"learning_rate": 4.114171633673705e-06,
|
| 687 |
+
"loss": 0.6238,
|
| 688 |
+
"step": 485
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.35251798561151076,
|
| 692 |
+
"grad_norm": 0.17043437687962124,
|
| 693 |
+
"learning_rate": 4.090074437942155e-06,
|
| 694 |
+
"loss": 0.6362,
|
| 695 |
+
"step": 490
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.35611510791366907,
|
| 699 |
+
"grad_norm": 0.17325913850748673,
|
| 700 |
+
"learning_rate": 4.065726552283173e-06,
|
| 701 |
+
"loss": 0.6567,
|
| 702 |
+
"step": 495
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3597122302158273,
|
| 706 |
+
"grad_norm": 0.19654711855055748,
|
| 707 |
+
"learning_rate": 4.0411318153659056e-06,
|
| 708 |
+
"loss": 0.6605,
|
| 709 |
+
"step": 500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3597122302158273,
|
| 713 |
+
"eval_runtime": 14.2753,
|
| 714 |
+
"eval_samples_per_second": 70.051,
|
| 715 |
+
"eval_steps_per_second": 2.242,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.36330935251798563,
|
| 720 |
+
"grad_norm": 0.181643640264244,
|
| 721 |
+
"learning_rate": 4.016294104777883e-06,
|
| 722 |
+
"loss": 0.6593,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3669064748201439,
|
| 727 |
+
"grad_norm": 0.17145091906098897,
|
| 728 |
+
"learning_rate": 3.99121733641368e-06,
|
| 729 |
+
"loss": 0.6434,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.37050359712230213,
|
| 734 |
+
"grad_norm": 0.18571422852307243,
|
| 735 |
+
"learning_rate": 3.96590546385754e-06,
|
| 736 |
+
"loss": 0.6351,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.37410071942446044,
|
| 741 |
+
"grad_norm": 0.1655963338752092,
|
| 742 |
+
"learning_rate": 3.9403624777600526e-06,
|
| 743 |
+
"loss": 0.6241,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3776978417266187,
|
| 748 |
+
"grad_norm": 0.21928102436797317,
|
| 749 |
+
"learning_rate": 3.914592405208993e-06,
|
| 750 |
+
"loss": 0.623,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.381294964028777,
|
| 755 |
+
"grad_norm": 0.17321888081042625,
|
| 756 |
+
"learning_rate": 3.888599309094413e-06,
|
| 757 |
+
"loss": 0.6289,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.38489208633093525,
|
| 762 |
+
"grad_norm": 0.17012813283361522,
|
| 763 |
+
"learning_rate": 3.862387287468095e-06,
|
| 764 |
+
"loss": 0.6268,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.38848920863309355,
|
| 769 |
+
"grad_norm": 0.1736732174250609,
|
| 770 |
+
"learning_rate": 3.835960472897444e-06,
|
| 771 |
+
"loss": 0.638,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3920863309352518,
|
| 776 |
+
"grad_norm": 0.23704552857732844,
|
| 777 |
+
"learning_rate": 3.809323031813963e-06,
|
| 778 |
+
"loss": 0.6244,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.39568345323741005,
|
| 783 |
+
"grad_norm": 0.16584229684045446,
|
| 784 |
+
"learning_rate": 3.7824791638563674e-06,
|
| 785 |
+
"loss": 0.6195,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.39928057553956836,
|
| 790 |
+
"grad_norm": 0.18088337678779626,
|
| 791 |
+
"learning_rate": 3.75543310120848e-06,
|
| 792 |
+
"loss": 0.6216,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.4028776978417266,
|
| 797 |
+
"grad_norm": 0.1670474782849822,
|
| 798 |
+
"learning_rate": 3.728189107931981e-06,
|
| 799 |
+
"loss": 0.629,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4064748201438849,
|
| 804 |
+
"grad_norm": 0.16881903430684905,
|
| 805 |
+
"learning_rate": 3.7007514792941462e-06,
|
| 806 |
+
"loss": 0.6159,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.41007194244604317,
|
| 811 |
+
"grad_norm": 0.166561835843837,
|
| 812 |
+
"learning_rate": 3.6731245410906537e-06,
|
| 813 |
+
"loss": 0.6182,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4136690647482014,
|
| 818 |
+
"grad_norm": 0.17513847235031135,
|
| 819 |
+
"learning_rate": 3.6453126489635845e-06,
|
| 820 |
+
"loss": 0.6469,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4172661870503597,
|
| 825 |
+
"grad_norm": 0.17220800442112816,
|
| 826 |
+
"learning_rate": 3.6173201877147134e-06,
|
| 827 |
+
"loss": 0.6342,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.420863309352518,
|
| 832 |
+
"grad_norm": 0.167483163354845,
|
| 833 |
+
"learning_rate": 3.5891515706142083e-06,
|
| 834 |
+
"loss": 0.6248,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4244604316546763,
|
| 839 |
+
"grad_norm": 0.16391025566890444,
|
| 840 |
+
"learning_rate": 3.560811238704832e-06,
|
| 841 |
+
"loss": 0.6285,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.42805755395683454,
|
| 846 |
+
"grad_norm": 0.17954405708333362,
|
| 847 |
+
"learning_rate": 3.532303660101776e-06,
|
| 848 |
+
"loss": 0.6653,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4316546762589928,
|
| 853 |
+
"grad_norm": 0.1748136517289543,
|
| 854 |
+
"learning_rate": 3.503633329288215e-06,
|
| 855 |
+
"loss": 0.61,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.4352517985611511,
|
| 860 |
+
"grad_norm": 0.16312734641366394,
|
| 861 |
+
"learning_rate": 3.474804766406718e-06,
|
| 862 |
+
"loss": 0.6507,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.43884892086330934,
|
| 867 |
+
"grad_norm": 0.16678530089886806,
|
| 868 |
+
"learning_rate": 3.445822516546598e-06,
|
| 869 |
+
"loss": 0.6391,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.44244604316546765,
|
| 874 |
+
"grad_norm": 0.16772897874751536,
|
| 875 |
+
"learning_rate": 3.416691149027341e-06,
|
| 876 |
+
"loss": 0.6488,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.4460431654676259,
|
| 881 |
+
"grad_norm": 0.17568340970424068,
|
| 882 |
+
"learning_rate": 3.3874152566782127e-06,
|
| 883 |
+
"loss": 0.6542,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.44964028776978415,
|
| 888 |
+
"grad_norm": 0.16904932894961983,
|
| 889 |
+
"learning_rate": 3.357999455114148e-06,
|
| 890 |
+
"loss": 0.6308,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.45323741007194246,
|
| 895 |
+
"grad_norm": 0.17964693226103262,
|
| 896 |
+
"learning_rate": 3.3284483820080694e-06,
|
| 897 |
+
"loss": 0.658,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.4568345323741007,
|
| 902 |
+
"grad_norm": 0.16972600698098864,
|
| 903 |
+
"learning_rate": 3.2987666963597006e-06,
|
| 904 |
+
"loss": 0.666,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.460431654676259,
|
| 909 |
+
"grad_norm": 0.16863410876839938,
|
| 910 |
+
"learning_rate": 3.2689590777610443e-06,
|
| 911 |
+
"loss": 0.6158,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.46402877697841727,
|
| 916 |
+
"grad_norm": 0.16752405859925734,
|
| 917 |
+
"learning_rate": 3.239030225658595e-06,
|
| 918 |
+
"loss": 0.6421,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4676258992805755,
|
| 923 |
+
"grad_norm": 0.16737880285691936,
|
| 924 |
+
"learning_rate": 3.208984858612429e-06,
|
| 925 |
+
"loss": 0.6567,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4712230215827338,
|
| 930 |
+
"grad_norm": 0.1622627800832757,
|
| 931 |
+
"learning_rate": 3.178827713552281e-06,
|
| 932 |
+
"loss": 0.6222,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.4748201438848921,
|
| 937 |
+
"grad_norm": 0.1637301148248093,
|
| 938 |
+
"learning_rate": 3.148563545030722e-06,
|
| 939 |
+
"loss": 0.6253,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4784172661870504,
|
| 944 |
+
"grad_norm": 0.1658915803907661,
|
| 945 |
+
"learning_rate": 3.1181971244735594e-06,
|
| 946 |
+
"loss": 0.6277,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.48201438848920863,
|
| 951 |
+
"grad_norm": 0.17005292953461104,
|
| 952 |
+
"learning_rate": 3.0877332394275806e-06,
|
| 953 |
+
"loss": 0.638,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4856115107913669,
|
| 958 |
+
"grad_norm": 0.17306211053376286,
|
| 959 |
+
"learning_rate": 3.05717669280575e-06,
|
| 960 |
+
"loss": 0.6317,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4892086330935252,
|
| 965 |
+
"grad_norm": 0.16464061138395658,
|
| 966 |
+
"learning_rate": 3.026532302129984e-06,
|
| 967 |
+
"loss": 0.6135,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.49280575539568344,
|
| 972 |
+
"grad_norm": 0.16817670931577094,
|
| 973 |
+
"learning_rate": 2.9958048987716266e-06,
|
| 974 |
+
"loss": 0.6255,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.49640287769784175,
|
| 979 |
+
"grad_norm": 0.16195389526156515,
|
| 980 |
+
"learning_rate": 2.96499932718974e-06,
|
| 981 |
+
"loss": 0.6024,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5,
|
| 986 |
+
"grad_norm": 0.1645172149025666,
|
| 987 |
+
"learning_rate": 2.9341204441673267e-06,
|
| 988 |
+
"loss": 0.6345,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5035971223021583,
|
| 993 |
+
"grad_norm": 0.22649243976417918,
|
| 994 |
+
"learning_rate": 2.903173118045616e-06,
|
| 995 |
+
"loss": 0.6312,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5071942446043165,
|
| 1000 |
+
"grad_norm": 0.17150190400496662,
|
| 1001 |
+
"learning_rate": 2.8721622279565248e-06,
|
| 1002 |
+
"loss": 0.642,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5107913669064749,
|
| 1007 |
+
"grad_norm": 0.16890621997495983,
|
| 1008 |
+
"learning_rate": 2.8410926630534164e-06,
|
| 1009 |
+
"loss": 0.6499,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5143884892086331,
|
| 1014 |
+
"grad_norm": 0.16876468012144627,
|
| 1015 |
+
"learning_rate": 2.8099693217402807e-06,
|
| 1016 |
+
"loss": 0.63,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5179856115107914,
|
| 1021 |
+
"grad_norm": 0.16959393931472438,
|
| 1022 |
+
"learning_rate": 2.7787971108994557e-06,
|
| 1023 |
+
"loss": 0.6181,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5215827338129496,
|
| 1028 |
+
"grad_norm": 0.17314847147328546,
|
| 1029 |
+
"learning_rate": 2.7475809451180103e-06,
|
| 1030 |
+
"loss": 0.6584,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5251798561151079,
|
| 1035 |
+
"grad_norm": 0.1687033450249176,
|
| 1036 |
+
"learning_rate": 2.7163257459129184e-06,
|
| 1037 |
+
"loss": 0.6308,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5287769784172662,
|
| 1042 |
+
"grad_norm": 0.1658057380008524,
|
| 1043 |
+
"learning_rate": 2.685036440955133e-06,
|
| 1044 |
+
"loss": 0.6529,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5323741007194245,
|
| 1049 |
+
"grad_norm": 0.16991236487195122,
|
| 1050 |
+
"learning_rate": 2.6537179632926953e-06,
|
| 1051 |
+
"loss": 0.6177,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5359712230215827,
|
| 1056 |
+
"grad_norm": 0.16035267630476027,
|
| 1057 |
+
"learning_rate": 2.6223752505729884e-06,
|
| 1058 |
+
"loss": 0.6461,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.539568345323741,
|
| 1063 |
+
"grad_norm": 0.16126651583706925,
|
| 1064 |
+
"learning_rate": 2.5910132442642815e-06,
|
| 1065 |
+
"loss": 0.6,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5431654676258992,
|
| 1070 |
+
"grad_norm": 0.1623864590113543,
|
| 1071 |
+
"learning_rate": 2.5596368888766537e-06,
|
| 1072 |
+
"loss": 0.5975,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5467625899280576,
|
| 1077 |
+
"grad_norm": 0.165811450449987,
|
| 1078 |
+
"learning_rate": 2.52825113118245e-06,
|
| 1079 |
+
"loss": 0.6292,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5503597122302158,
|
| 1084 |
+
"grad_norm": 0.16532836548055116,
|
| 1085 |
+
"learning_rate": 2.496860919436374e-06,
|
| 1086 |
+
"loss": 0.6084,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5539568345323741,
|
| 1091 |
+
"grad_norm": 0.16785794412521857,
|
| 1092 |
+
"learning_rate": 2.4654712025953543e-06,
|
| 1093 |
+
"loss": 0.6087,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5575539568345323,
|
| 1098 |
+
"grad_norm": 0.1635851401173821,
|
| 1099 |
+
"learning_rate": 2.4340869295382924e-06,
|
| 1100 |
+
"loss": 0.6328,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5611510791366906,
|
| 1105 |
+
"grad_norm": 0.1600778829109294,
|
| 1106 |
+
"learning_rate": 2.402713048285825e-06,
|
| 1107 |
+
"loss": 0.6033,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.564748201438849,
|
| 1112 |
+
"grad_norm": 0.18588185959536366,
|
| 1113 |
+
"learning_rate": 2.3713545052202268e-06,
|
| 1114 |
+
"loss": 0.6158,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5683453237410072,
|
| 1119 |
+
"grad_norm": 0.16421497309958943,
|
| 1120 |
+
"learning_rate": 2.3400162443055655e-06,
|
| 1121 |
+
"loss": 0.604,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5719424460431655,
|
| 1126 |
+
"grad_norm": 0.160265450599092,
|
| 1127 |
+
"learning_rate": 2.30870320630824e-06,
|
| 1128 |
+
"loss": 0.612,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5755395683453237,
|
| 1133 |
+
"grad_norm": 0.16954798697965026,
|
| 1134 |
+
"learning_rate": 2.277420328018023e-06,
|
| 1135 |
+
"loss": 0.6366,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.579136690647482,
|
| 1140 |
+
"grad_norm": 0.1611435816780583,
|
| 1141 |
+
"learning_rate": 2.24617254146973e-06,
|
| 1142 |
+
"loss": 0.6349,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5827338129496403,
|
| 1147 |
+
"grad_norm": 0.17546659615139334,
|
| 1148 |
+
"learning_rate": 2.214964773165641e-06,
|
| 1149 |
+
"loss": 0.5949,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5863309352517986,
|
| 1154 |
+
"grad_norm": 0.16049134388976258,
|
| 1155 |
+
"learning_rate": 2.183801943298789e-06,
|
| 1156 |
+
"loss": 0.6067,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5899280575539568,
|
| 1161 |
+
"grad_norm": 0.16957473530909392,
|
| 1162 |
+
"learning_rate": 2.1526889649772477e-06,
|
| 1163 |
+
"loss": 0.5934,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5935251798561151,
|
| 1168 |
+
"grad_norm": 0.16581052754385936,
|
| 1169 |
+
"learning_rate": 2.121630743449532e-06,
|
| 1170 |
+
"loss": 0.6171,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5971223021582733,
|
| 1175 |
+
"grad_norm": 0.16120322904660017,
|
| 1176 |
+
"learning_rate": 2.090632175331244e-06,
|
| 1177 |
+
"loss": 0.6163,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6007194244604317,
|
| 1182 |
+
"grad_norm": 0.1581771997504381,
|
| 1183 |
+
"learning_rate": 2.059698147833075e-06,
|
| 1184 |
+
"loss": 0.6197,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.60431654676259,
|
| 1189 |
+
"grad_norm": 0.16362498684959523,
|
| 1190 |
+
"learning_rate": 2.0288335379902895e-06,
|
| 1191 |
+
"loss": 0.6207,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6079136690647482,
|
| 1196 |
+
"grad_norm": 0.16660059907607108,
|
| 1197 |
+
"learning_rate": 1.9980432118938204e-06,
|
| 1198 |
+
"loss": 0.6354,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6115107913669064,
|
| 1203 |
+
"grad_norm": 0.1803995538499011,
|
| 1204 |
+
"learning_rate": 1.9673320239230783e-06,
|
| 1205 |
+
"loss": 0.6041,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6151079136690647,
|
| 1210 |
+
"grad_norm": 0.20020108799514952,
|
| 1211 |
+
"learning_rate": 1.9367048159806175e-06,
|
| 1212 |
+
"loss": 0.6236,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6187050359712231,
|
| 1217 |
+
"grad_norm": 0.16682386811464148,
|
| 1218 |
+
"learning_rate": 1.9061664167287672e-06,
|
| 1219 |
+
"loss": 0.6242,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6223021582733813,
|
| 1224 |
+
"grad_norm": 0.1637761383770433,
|
| 1225 |
+
"learning_rate": 1.875721640828344e-06,
|
| 1226 |
+
"loss": 0.6039,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6258992805755396,
|
| 1231 |
+
"grad_norm": 0.1624753055536377,
|
| 1232 |
+
"learning_rate": 1.8453752881795772e-06,
|
| 1233 |
+
"loss": 0.612,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6294964028776978,
|
| 1238 |
+
"grad_norm": 0.16504744819300535,
|
| 1239 |
+
"learning_rate": 1.8151321431653627e-06,
|
| 1240 |
+
"loss": 0.6071,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6330935251798561,
|
| 1245 |
+
"grad_norm": 0.16453571982489182,
|
| 1246 |
+
"learning_rate": 1.7849969738969592e-06,
|
| 1247 |
+
"loss": 0.6163,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6366906474820144,
|
| 1252 |
+
"grad_norm": 0.185104019176102,
|
| 1253 |
+
"learning_rate": 1.754974531462251e-06,
|
| 1254 |
+
"loss": 0.6329,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6402877697841727,
|
| 1259 |
+
"grad_norm": 0.1677831386628234,
|
| 1260 |
+
"learning_rate": 1.725069549176695e-06,
|
| 1261 |
+
"loss": 0.6046,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6438848920863309,
|
| 1266 |
+
"grad_norm": 0.1585666383654991,
|
| 1267 |
+
"learning_rate": 1.6952867418370707e-06,
|
| 1268 |
+
"loss": 0.6002,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6474820143884892,
|
| 1273 |
+
"grad_norm": 0.16489693303482242,
|
| 1274 |
+
"learning_rate": 1.665630804978149e-06,
|
| 1275 |
+
"loss": 0.6376,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6510791366906474,
|
| 1280 |
+
"grad_norm": 0.16805125828505416,
|
| 1281 |
+
"learning_rate": 1.6361064141323953e-06,
|
| 1282 |
+
"loss": 0.6365,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6546762589928058,
|
| 1287 |
+
"grad_norm": 0.181566998160182,
|
| 1288 |
+
"learning_rate": 1.6067182240928332e-06,
|
| 1289 |
+
"loss": 0.6112,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.658273381294964,
|
| 1294 |
+
"grad_norm": 0.1694206935808062,
|
| 1295 |
+
"learning_rate": 1.5774708681791692e-06,
|
| 1296 |
+
"loss": 0.6187,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6618705035971223,
|
| 1301 |
+
"grad_norm": 0.16461060571783928,
|
| 1302 |
+
"learning_rate": 1.548368957507308e-06,
|
| 1303 |
+
"loss": 0.5782,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6654676258992805,
|
| 1308 |
+
"grad_norm": 0.1669683477972491,
|
| 1309 |
+
"learning_rate": 1.5194170802623692e-06,
|
| 1310 |
+
"loss": 0.6145,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6690647482014388,
|
| 1315 |
+
"grad_norm": 0.16862830642330334,
|
| 1316 |
+
"learning_rate": 1.4906198009753159e-06,
|
| 1317 |
+
"loss": 0.6353,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6726618705035972,
|
| 1322 |
+
"grad_norm": 0.17094178135804486,
|
| 1323 |
+
"learning_rate": 1.4619816598033148e-06,
|
| 1324 |
+
"loss": 0.597,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6762589928057554,
|
| 1329 |
+
"grad_norm": 0.1594405984136636,
|
| 1330 |
+
"learning_rate": 1.4335071718139379e-06,
|
| 1331 |
+
"loss": 0.6141,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6798561151079137,
|
| 1336 |
+
"grad_norm": 0.16333147484898983,
|
| 1337 |
+
"learning_rate": 1.4052008262733205e-06,
|
| 1338 |
+
"loss": 0.6048,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6834532374100719,
|
| 1343 |
+
"grad_norm": 0.15979882139296125,
|
| 1344 |
+
"learning_rate": 1.3770670859383895e-06,
|
| 1345 |
+
"loss": 0.6217,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6870503597122302,
|
| 1350 |
+
"grad_norm": 0.16504220868699096,
|
| 1351 |
+
"learning_rate": 1.3491103863532626e-06,
|
| 1352 |
+
"loss": 0.6052,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6906474820143885,
|
| 1357 |
+
"grad_norm": 0.17583165256569067,
|
| 1358 |
+
"learning_rate": 1.321335135149952e-06,
|
| 1359 |
+
"loss": 0.6309,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6942446043165468,
|
| 1364 |
+
"grad_norm": 0.18291087991427973,
|
| 1365 |
+
"learning_rate": 1.2937457113534498e-06,
|
| 1366 |
+
"loss": 0.6152,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.697841726618705,
|
| 1371 |
+
"grad_norm": 0.15817294651209343,
|
| 1372 |
+
"learning_rate": 1.266346464691346e-06,
|
| 1373 |
+
"loss": 0.6186,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7014388489208633,
|
| 1378 |
+
"grad_norm": 0.16487935933779904,
|
| 1379 |
+
"learning_rate": 1.2391417149080458e-06,
|
| 1380 |
+
"loss": 0.6206,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7050359712230215,
|
| 1385 |
+
"grad_norm": 0.16501096888220573,
|
| 1386 |
+
"learning_rate": 1.212135751083724e-06,
|
| 1387 |
+
"loss": 0.642,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.7086330935251799,
|
| 1392 |
+
"grad_norm": 0.16345813795019531,
|
| 1393 |
+
"learning_rate": 1.1853328309581139e-06,
|
| 1394 |
+
"loss": 0.6105,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7122302158273381,
|
| 1399 |
+
"grad_norm": 0.16982887708707609,
|
| 1400 |
+
"learning_rate": 1.1587371802592302e-06,
|
| 1401 |
+
"loss": 0.6302,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7158273381294964,
|
| 1406 |
+
"grad_norm": 0.15843844245859565,
|
| 1407 |
+
"learning_rate": 1.1323529920371518e-06,
|
| 1408 |
+
"loss": 0.5946,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7194244604316546,
|
| 1413 |
+
"grad_norm": 0.16408266793537263,
|
| 1414 |
+
"learning_rate": 1.10618442600294e-06,
|
| 1415 |
+
"loss": 0.6108,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7194244604316546,
|
| 1420 |
+
"eval_runtime": 12.2588,
|
| 1421 |
+
"eval_samples_per_second": 81.574,
|
| 1422 |
+
"eval_steps_per_second": 2.61,
|
| 1423 |
+
"step": 1000
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.7230215827338129,
|
| 1427 |
+
"grad_norm": 0.16405449199837346,
|
| 1428 |
+
"learning_rate": 1.0802356078728293e-06,
|
| 1429 |
+
"loss": 0.6247,
|
| 1430 |
+
"step": 1005
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7266187050359713,
|
| 1434 |
+
"grad_norm": 0.17021168722767338,
|
| 1435 |
+
"learning_rate": 1.0545106287177645e-06,
|
| 1436 |
+
"loss": 0.6146,
|
| 1437 |
+
"step": 1010
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.7302158273381295,
|
| 1441 |
+
"grad_norm": 0.1663335063999724,
|
| 1442 |
+
"learning_rate": 1.029013544318407e-06,
|
| 1443 |
+
"loss": 0.6128,
|
| 1444 |
+
"step": 1015
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7338129496402878,
|
| 1448 |
+
"grad_norm": 0.16192750797174188,
|
| 1449 |
+
"learning_rate": 1.0037483745257073e-06,
|
| 1450 |
+
"loss": 0.6282,
|
| 1451 |
+
"step": 1020
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.737410071942446,
|
| 1455 |
+
"grad_norm": 0.1676690522330883,
|
| 1456 |
+
"learning_rate": 9.78719102627132e-07,
|
| 1457 |
+
"loss": 0.6405,
|
| 1458 |
+
"step": 1025
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7410071942446043,
|
| 1462 |
+
"grad_norm": 0.16103562098112373,
|
| 1463 |
+
"learning_rate": 9.53929674718668e-07,
|
| 1464 |
+
"loss": 0.6092,
|
| 1465 |
+
"step": 1030
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7446043165467626,
|
| 1469 |
+
"grad_norm": 0.15629957097839647,
|
| 1470 |
+
"learning_rate": 9.293839990826778e-07,
|
| 1471 |
+
"loss": 0.6288,
|
| 1472 |
+
"step": 1035
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7482014388489209,
|
| 1476 |
+
"grad_norm": 0.1649719012991602,
|
| 1477 |
+
"learning_rate": 9.050859455717292e-07,
|
| 1478 |
+
"loss": 0.6487,
|
| 1479 |
+
"step": 1040
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7517985611510791,
|
| 1483 |
+
"grad_norm": 0.16621814957301698,
|
| 1484 |
+
"learning_rate": 8.810393449984706e-07,
|
| 1485 |
+
"loss": 0.6579,
|
| 1486 |
+
"step": 1045
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7553956834532374,
|
| 1490 |
+
"grad_norm": 0.15499205580375533,
|
| 1491 |
+
"learning_rate": 8.572479885316745e-07,
|
| 1492 |
+
"loss": 0.6113,
|
| 1493 |
+
"step": 1050
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7589928057553957,
|
| 1497 |
+
"grad_norm": 0.16101254180295285,
|
| 1498 |
+
"learning_rate": 8.337156270985197e-07,
|
| 1499 |
+
"loss": 0.6414,
|
| 1500 |
+
"step": 1055
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.762589928057554,
|
| 1504 |
+
"grad_norm": 0.16549776608782082,
|
| 1505 |
+
"learning_rate": 8.104459707932238e-07,
|
| 1506 |
+
"loss": 0.6123,
|
| 1507 |
+
"step": 1060
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7661870503597122,
|
| 1511 |
+
"grad_norm": 0.17260224506613606,
|
| 1512 |
+
"learning_rate": 7.874426882921171e-07,
|
| 1513 |
+
"loss": 0.6274,
|
| 1514 |
+
"step": 1065
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7697841726618705,
|
| 1518 |
+
"grad_norm": 0.16949981491490823,
|
| 1519 |
+
"learning_rate": 7.647094062752347e-07,
|
| 1520 |
+
"loss": 0.6253,
|
| 1521 |
+
"step": 1070
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.7733812949640287,
|
| 1525 |
+
"grad_norm": 0.16631605946143985,
|
| 1526 |
+
"learning_rate": 7.422497088545436e-07,
|
| 1527 |
+
"loss": 0.6233,
|
| 1528 |
+
"step": 1075
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7769784172661871,
|
| 1532 |
+
"grad_norm": 0.16504960475084282,
|
| 1533 |
+
"learning_rate": 7.200671370088682e-07,
|
| 1534 |
+
"loss": 0.5962,
|
| 1535 |
+
"step": 1080
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7805755395683454,
|
| 1539 |
+
"grad_norm": 0.15889073547280716,
|
| 1540 |
+
"learning_rate": 6.981651880256285e-07,
|
| 1541 |
+
"loss": 0.6094,
|
| 1542 |
+
"step": 1085
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7841726618705036,
|
| 1546 |
+
"grad_norm": 0.15921735320612743,
|
| 1547 |
+
"learning_rate": 6.765473149494545e-07,
|
| 1548 |
+
"loss": 0.6271,
|
| 1549 |
+
"step": 1090
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7877697841726619,
|
| 1553 |
+
"grad_norm": 0.16149400411033712,
|
| 1554 |
+
"learning_rate": 6.552169260377872e-07,
|
| 1555 |
+
"loss": 0.6194,
|
| 1556 |
+
"step": 1095
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7913669064748201,
|
| 1560 |
+
"grad_norm": 0.17519197732403996,
|
| 1561 |
+
"learning_rate": 6.341773842235307e-07,
|
| 1562 |
+
"loss": 0.5996,
|
| 1563 |
+
"step": 1100
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7949640287769785,
|
| 1567 |
+
"grad_norm": 0.16208379962765676,
|
| 1568 |
+
"learning_rate": 6.134320065848564e-07,
|
| 1569 |
+
"loss": 0.606,
|
| 1570 |
+
"step": 1105
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7985611510791367,
|
| 1574 |
+
"grad_norm": 0.17454197491728474,
|
| 1575 |
+
"learning_rate": 5.929840638222384e-07,
|
| 1576 |
+
"loss": 0.5922,
|
| 1577 |
+
"step": 1110
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.802158273381295,
|
| 1581 |
+
"grad_norm": 0.1646154768218277,
|
| 1582 |
+
"learning_rate": 5.728367797427906e-07,
|
| 1583 |
+
"loss": 0.6243,
|
| 1584 |
+
"step": 1115
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8057553956834532,
|
| 1588 |
+
"grad_norm": 0.16381184894925108,
|
| 1589 |
+
"learning_rate": 5.529933307520102e-07,
|
| 1590 |
+
"loss": 0.591,
|
| 1591 |
+
"step": 1120
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8093525179856115,
|
| 1595 |
+
"grad_norm": 0.16087653845215716,
|
| 1596 |
+
"learning_rate": 5.334568453529831e-07,
|
| 1597 |
+
"loss": 0.6032,
|
| 1598 |
+
"step": 1125
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8129496402877698,
|
| 1602 |
+
"grad_norm": 0.169499547501472,
|
| 1603 |
+
"learning_rate": 5.142304036531503e-07,
|
| 1604 |
+
"loss": 0.6272,
|
| 1605 |
+
"step": 1130
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.8165467625899281,
|
| 1609 |
+
"grad_norm": 0.16073871102203974,
|
| 1610 |
+
"learning_rate": 4.953170368786985e-07,
|
| 1611 |
+
"loss": 0.6038,
|
| 1612 |
+
"step": 1135
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8201438848920863,
|
| 1616 |
+
"grad_norm": 0.1718082418985466,
|
| 1617 |
+
"learning_rate": 4.767197268966589e-07,
|
| 1618 |
+
"loss": 0.6329,
|
| 1619 |
+
"step": 1140
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.8237410071942446,
|
| 1623 |
+
"grad_norm": 0.16600964986098235,
|
| 1624 |
+
"learning_rate": 4.5844140574478944e-07,
|
| 1625 |
+
"loss": 0.6059,
|
| 1626 |
+
"step": 1145
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8273381294964028,
|
| 1630 |
+
"grad_norm": 0.16551060707936915,
|
| 1631 |
+
"learning_rate": 4.404849551693102e-07,
|
| 1632 |
+
"loss": 0.6178,
|
| 1633 |
+
"step": 1150
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.8309352517985612,
|
| 1637 |
+
"grad_norm": 0.16058229980877273,
|
| 1638 |
+
"learning_rate": 4.228532061705742e-07,
|
| 1639 |
+
"loss": 0.6162,
|
| 1640 |
+
"step": 1155
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.8345323741007195,
|
| 1644 |
+
"grad_norm": 0.1650296995501976,
|
| 1645 |
+
"learning_rate": 4.055489385567266e-07,
|
| 1646 |
+
"loss": 0.6252,
|
| 1647 |
+
"step": 1160
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.8381294964028777,
|
| 1651 |
+
"grad_norm": 0.16480409428538245,
|
| 1652 |
+
"learning_rate": 3.8857488050544903e-07,
|
| 1653 |
+
"loss": 0.6203,
|
| 1654 |
+
"step": 1165
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.841726618705036,
|
| 1658 |
+
"grad_norm": 0.16232443768326824,
|
| 1659 |
+
"learning_rate": 3.7193370813383425e-07,
|
| 1660 |
+
"loss": 0.6349,
|
| 1661 |
+
"step": 1170
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8453237410071942,
|
| 1665 |
+
"grad_norm": 0.1584897308647271,
|
| 1666 |
+
"learning_rate": 3.556280450764699e-07,
|
| 1667 |
+
"loss": 0.584,
|
| 1668 |
+
"step": 1175
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8489208633093526,
|
| 1672 |
+
"grad_norm": 0.15664466526812762,
|
| 1673 |
+
"learning_rate": 3.396604620718025e-07,
|
| 1674 |
+
"loss": 0.5986,
|
| 1675 |
+
"step": 1180
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8525179856115108,
|
| 1679 |
+
"grad_norm": 4.159707417477735,
|
| 1680 |
+
"learning_rate": 3.240334765568329e-07,
|
| 1681 |
+
"loss": 0.6446,
|
| 1682 |
+
"step": 1185
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8561151079136691,
|
| 1686 |
+
"grad_norm": 0.15382670213595884,
|
| 1687 |
+
"learning_rate": 3.0874955227022053e-07,
|
| 1688 |
+
"loss": 0.6192,
|
| 1689 |
+
"step": 1190
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8597122302158273,
|
| 1693 |
+
"grad_norm": 0.16333556616379255,
|
| 1694 |
+
"learning_rate": 2.938110988638521e-07,
|
| 1695 |
+
"loss": 0.5958,
|
| 1696 |
+
"step": 1195
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8633093525179856,
|
| 1700 |
+
"grad_norm": 0.1590368907083947,
|
| 1701 |
+
"learning_rate": 2.7922047152293934e-07,
|
| 1702 |
+
"loss": 0.6197,
|
| 1703 |
+
"step": 1200
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8669064748201439,
|
| 1707 |
+
"grad_norm": 0.17020833161122917,
|
| 1708 |
+
"learning_rate": 2.6497997059470065e-07,
|
| 1709 |
+
"loss": 0.5887,
|
| 1710 |
+
"step": 1205
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8705035971223022,
|
| 1714 |
+
"grad_norm": 0.16037027743271673,
|
| 1715 |
+
"learning_rate": 2.5109184122568797e-07,
|
| 1716 |
+
"loss": 0.5971,
|
| 1717 |
+
"step": 1210
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8741007194244604,
|
| 1721 |
+
"grad_norm": 0.1633515403590847,
|
| 1722 |
+
"learning_rate": 2.3755827300782436e-07,
|
| 1723 |
+
"loss": 0.6147,
|
| 1724 |
+
"step": 1215
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8776978417266187,
|
| 1728 |
+
"grad_norm": 0.15724633669451,
|
| 1729 |
+
"learning_rate": 2.2438139963318812e-07,
|
| 1730 |
+
"loss": 0.5837,
|
| 1731 |
+
"step": 1220
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8812949640287769,
|
| 1735 |
+
"grad_norm": 0.1606820436871867,
|
| 1736 |
+
"learning_rate": 2.1156329855762243e-07,
|
| 1737 |
+
"loss": 0.6084,
|
| 1738 |
+
"step": 1225
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8848920863309353,
|
| 1742 |
+
"grad_norm": 0.1665829512742088,
|
| 1743 |
+
"learning_rate": 1.9910599067319984e-07,
|
| 1744 |
+
"loss": 0.5792,
|
| 1745 |
+
"step": 1230
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8884892086330936,
|
| 1749 |
+
"grad_norm": 0.1617341090573433,
|
| 1750 |
+
"learning_rate": 1.8701143998961502e-07,
|
| 1751 |
+
"loss": 0.6127,
|
| 1752 |
+
"step": 1235
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8920863309352518,
|
| 1756 |
+
"grad_norm": 0.16444482030532112,
|
| 1757 |
+
"learning_rate": 1.752815533245364e-07,
|
| 1758 |
+
"loss": 0.6068,
|
| 1759 |
+
"step": 1240
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.89568345323741,
|
| 1763 |
+
"grad_norm": 0.16828126570381266,
|
| 1764 |
+
"learning_rate": 1.6391818000298043e-07,
|
| 1765 |
+
"loss": 0.6125,
|
| 1766 |
+
"step": 1245
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8992805755395683,
|
| 1770 |
+
"grad_norm": 0.1629957454088181,
|
| 1771 |
+
"learning_rate": 1.529231115657498e-07,
|
| 1772 |
+
"loss": 0.6381,
|
| 1773 |
+
"step": 1250
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.9028776978417267,
|
| 1777 |
+
"grad_norm": 0.16556649184516456,
|
| 1778 |
+
"learning_rate": 1.4229808148697732e-07,
|
| 1779 |
+
"loss": 0.5998,
|
| 1780 |
+
"step": 1255
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9064748201438849,
|
| 1784 |
+
"grad_norm": 0.16203099377866306,
|
| 1785 |
+
"learning_rate": 1.320447649008308e-07,
|
| 1786 |
+
"loss": 0.6137,
|
| 1787 |
+
"step": 1260
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9100719424460432,
|
| 1791 |
+
"grad_norm": 0.16902179185053035,
|
| 1792 |
+
"learning_rate": 1.2216477833741025e-07,
|
| 1793 |
+
"loss": 0.5974,
|
| 1794 |
+
"step": 1265
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.9136690647482014,
|
| 1798 |
+
"grad_norm": 0.16479092334567338,
|
| 1799 |
+
"learning_rate": 1.1265967946788913e-07,
|
| 1800 |
+
"loss": 0.6203,
|
| 1801 |
+
"step": 1270
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.9172661870503597,
|
| 1805 |
+
"grad_norm": 0.15887060324938326,
|
| 1806 |
+
"learning_rate": 1.0353096685893044e-07,
|
| 1807 |
+
"loss": 0.5961,
|
| 1808 |
+
"step": 1275
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.920863309352518,
|
| 1812 |
+
"grad_norm": 0.15717928347327234,
|
| 1813 |
+
"learning_rate": 9.478007973642733e-08,
|
| 1814 |
+
"loss": 0.5939,
|
| 1815 |
+
"step": 1280
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.9244604316546763,
|
| 1819 |
+
"grad_norm": 0.16321773056840833,
|
| 1820 |
+
"learning_rate": 8.640839775859222e-08,
|
| 1821 |
+
"loss": 0.5993,
|
| 1822 |
+
"step": 1285
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9280575539568345,
|
| 1826 |
+
"grad_norm": 0.15772198437541257,
|
| 1827 |
+
"learning_rate": 7.841724079844321e-08,
|
| 1828 |
+
"loss": 0.6423,
|
| 1829 |
+
"step": 1290
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.9316546762589928,
|
| 1833 |
+
"grad_norm": 0.16917268204241048,
|
| 1834 |
+
"learning_rate": 7.080786873571388e-08,
|
| 1835 |
+
"loss": 0.6566,
|
| 1836 |
+
"step": 1295
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.935251798561151,
|
| 1840 |
+
"grad_norm": 0.16079832958818444,
|
| 1841 |
+
"learning_rate": 6.358148125822e-08,
|
| 1842 |
+
"loss": 0.6154,
|
| 1843 |
+
"step": 1300
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.9388489208633094,
|
| 1847 |
+
"grad_norm": 0.15308750470330593,
|
| 1848 |
+
"learning_rate": 5.673921767271967e-08,
|
| 1849 |
+
"loss": 0.6068,
|
| 1850 |
+
"step": 1305
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.9424460431654677,
|
| 1854 |
+
"grad_norm": 0.1675136373579217,
|
| 1855 |
+
"learning_rate": 5.028215672528924e-08,
|
| 1856 |
+
"loss": 0.6031,
|
| 1857 |
+
"step": 1310
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9460431654676259,
|
| 1861 |
+
"grad_norm": 0.15773902663760664,
|
| 1862 |
+
"learning_rate": 4.421131643125104e-08,
|
| 1863 |
+
"loss": 0.5913,
|
| 1864 |
+
"step": 1315
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.9496402877697842,
|
| 1868 |
+
"grad_norm": 0.15738955399333943,
|
| 1869 |
+
"learning_rate": 3.852765391467117e-08,
|
| 1870 |
+
"loss": 0.6246,
|
| 1871 |
+
"step": 1320
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9532374100719424,
|
| 1875 |
+
"grad_norm": 0.1673309275263978,
|
| 1876 |
+
"learning_rate": 3.323206525746303e-08,
|
| 1877 |
+
"loss": 0.6113,
|
| 1878 |
+
"step": 1325
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.9568345323741008,
|
| 1882 |
+
"grad_norm": 0.16639511680642716,
|
| 1883 |
+
"learning_rate": 2.832538535810947e-08,
|
| 1884 |
+
"loss": 0.6129,
|
| 1885 |
+
"step": 1330
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.960431654676259,
|
| 1889 |
+
"grad_norm": 0.15467986329263658,
|
| 1890 |
+
"learning_rate": 2.3808387800034194e-08,
|
| 1891 |
+
"loss": 0.6135,
|
| 1892 |
+
"step": 1335
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9640287769784173,
|
| 1896 |
+
"grad_norm": 0.15556453745576904,
|
| 1897 |
+
"learning_rate": 1.9681784729639608e-08,
|
| 1898 |
+
"loss": 0.5911,
|
| 1899 |
+
"step": 1340
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9676258992805755,
|
| 1903 |
+
"grad_norm": 0.17175964565211488,
|
| 1904 |
+
"learning_rate": 1.5946226744029402e-08,
|
| 1905 |
+
"loss": 0.6353,
|
| 1906 |
+
"step": 1345
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9712230215827338,
|
| 1910 |
+
"grad_norm": 0.15611077883391353,
|
| 1911 |
+
"learning_rate": 1.2602302788436715e-08,
|
| 1912 |
+
"loss": 0.5965,
|
| 1913 |
+
"step": 1350
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.9748201438848921,
|
| 1917 |
+
"grad_norm": 0.15835204960049692,
|
| 1918 |
+
"learning_rate": 9.650540063370628e-09,
|
| 1919 |
+
"loss": 0.602,
|
| 1920 |
+
"step": 1355
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9784172661870504,
|
| 1924 |
+
"grad_norm": 0.16328945271739753,
|
| 1925 |
+
"learning_rate": 7.091403941499597e-09,
|
| 1926 |
+
"loss": 0.5979,
|
| 1927 |
+
"step": 1360
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.9820143884892086,
|
| 1931 |
+
"grad_norm": 0.1583518161160204,
|
| 1932 |
+
"learning_rate": 4.9252978942793125e-09,
|
| 1933 |
+
"loss": 0.5995,
|
| 1934 |
+
"step": 1365
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.9856115107913669,
|
| 1938 |
+
"grad_norm": 0.16148337835743226,
|
| 1939 |
+
"learning_rate": 3.152563428343025e-09,
|
| 1940 |
+
"loss": 0.5697,
|
| 1941 |
+
"step": 1370
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9892086330935251,
|
| 1945 |
+
"grad_norm": 0.16118815430555725,
|
| 1946 |
+
"learning_rate": 1.7734800316596135e-09,
|
| 1947 |
+
"loss": 0.6051,
|
| 1948 |
+
"step": 1375
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9928057553956835,
|
| 1952 |
+
"grad_norm": 0.1572343438101448,
|
| 1953 |
+
"learning_rate": 7.882651294685573e-10,
|
| 1954 |
+
"loss": 0.6255,
|
| 1955 |
+
"step": 1380
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.9964028776978417,
|
| 1959 |
+
"grad_norm": 0.16657064617868036,
|
| 1960 |
+
"learning_rate": 1.9707405000346513e-10,
|
| 1961 |
+
"loss": 0.6444,
|
| 1962 |
+
"step": 1385
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 1.0,
|
| 1966 |
+
"grad_norm": 0.15928289801803977,
|
| 1967 |
+
"learning_rate": 0.0,
|
| 1968 |
+
"loss": 0.5923,
|
| 1969 |
+
"step": 1390
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 1.0,
|
| 1973 |
+
"step": 1390,
|
| 1974 |
+
"total_flos": 1267133463920640.0,
|
| 1975 |
+
"train_loss": 0.6670485469077131,
|
| 1976 |
+
"train_runtime": 18244.3099,
|
| 1977 |
+
"train_samples_per_second": 9.746,
|
| 1978 |
+
"train_steps_per_second": 0.076
|
| 1979 |
+
}
|
| 1980 |
+
],
|
| 1981 |
+
"logging_steps": 5,
|
| 1982 |
+
"max_steps": 1390,
|
| 1983 |
+
"num_input_tokens_seen": 0,
|
| 1984 |
+
"num_train_epochs": 1,
|
| 1985 |
+
"save_steps": 50,
|
| 1986 |
+
"stateful_callbacks": {
|
| 1987 |
+
"TrainerControl": {
|
| 1988 |
+
"args": {
|
| 1989 |
+
"should_epoch_stop": false,
|
| 1990 |
+
"should_evaluate": false,
|
| 1991 |
+
"should_log": false,
|
| 1992 |
+
"should_save": true,
|
| 1993 |
+
"should_training_stop": true
|
| 1994 |
+
},
|
| 1995 |
+
"attributes": {}
|
| 1996 |
+
}
|
| 1997 |
+
},
|
| 1998 |
+
"total_flos": 1267133463920640.0,
|
| 1999 |
+
"train_batch_size": 32,
|
| 2000 |
+
"trial_name": null,
|
| 2001 |
+
"trial_params": null
|
| 2002 |
+
}
|