Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- generation_config.json +9 -0
- train_results.json +9 -0
- trainer_state.json +2717 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Llama-3.1-8B
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Llama-3.1-8B-sft-ultrachat
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Llama-3.1-8B-sft-ultrachat
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B).
|
| 15 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
+
|
| 17 |
+
## Quick start
|
| 18 |
+
|
| 19 |
+
```python
|
| 20 |
+
from transformers import pipeline
|
| 21 |
+
|
| 22 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 23 |
+
generator = pipeline("text-generation", model="AmberYifan/Llama-3.1-8B-sft-ultrachat", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.2
|
| 37 |
+
- Transformers: 4.46.3
|
| 38 |
+
- Pytorch: 2.5.1+cu118
|
| 39 |
+
- Datasets: 3.2.0
|
| 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 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 99455336448000.0,
|
| 4 |
+
"train_loss": 1.1960604431754664,
|
| 5 |
+
"train_runtime": 10060.3773,
|
| 6 |
+
"train_samples": 51966,
|
| 7 |
+
"train_samples_per_second": 6.043,
|
| 8 |
+
"train_steps_per_second": 0.189
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": 128001,
|
| 6 |
+
"temperature": 0.6,
|
| 7 |
+
"top_p": 0.9,
|
| 8 |
+
"transformers_version": "4.46.3"
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 99455336448000.0,
|
| 4 |
+
"train_loss": 1.1960604431754664,
|
| 5 |
+
"train_runtime": 10060.3773,
|
| 6 |
+
"train_samples": 51966,
|
| 7 |
+
"train_samples_per_second": 6.043,
|
| 8 |
+
"train_steps_per_second": 0.189
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2717 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1900,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005263157894736842,
|
| 13 |
+
"grad_norm": 17.523106343551206,
|
| 14 |
+
"learning_rate": 1.0526315789473685e-07,
|
| 15 |
+
"loss": 1.5333,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.002631578947368421,
|
| 20 |
+
"grad_norm": 18.121909431354247,
|
| 21 |
+
"learning_rate": 5.263157894736843e-07,
|
| 22 |
+
"loss": 1.5177,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005263157894736842,
|
| 27 |
+
"grad_norm": 14.836011770372137,
|
| 28 |
+
"learning_rate": 1.0526315789473685e-06,
|
| 29 |
+
"loss": 1.4389,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.007894736842105263,
|
| 34 |
+
"grad_norm": 3.8440549265416597,
|
| 35 |
+
"learning_rate": 1.5789473684210526e-06,
|
| 36 |
+
"loss": 1.4167,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010526315789473684,
|
| 41 |
+
"grad_norm": 2.4276964194723165,
|
| 42 |
+
"learning_rate": 2.105263157894737e-06,
|
| 43 |
+
"loss": 1.2742,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013157894736842105,
|
| 48 |
+
"grad_norm": 2.187369889390014,
|
| 49 |
+
"learning_rate": 2.631578947368421e-06,
|
| 50 |
+
"loss": 1.231,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.015789473684210527,
|
| 55 |
+
"grad_norm": 2.0926997492754285,
|
| 56 |
+
"learning_rate": 3.157894736842105e-06,
|
| 57 |
+
"loss": 1.2264,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.018421052631578946,
|
| 62 |
+
"grad_norm": 2.0987185493937197,
|
| 63 |
+
"learning_rate": 3.6842105263157896e-06,
|
| 64 |
+
"loss": 1.1576,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.021052631578947368,
|
| 69 |
+
"grad_norm": 1.871475479211144,
|
| 70 |
+
"learning_rate": 4.210526315789474e-06,
|
| 71 |
+
"loss": 1.2529,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02368421052631579,
|
| 76 |
+
"grad_norm": 2.2193852291586773,
|
| 77 |
+
"learning_rate": 4.736842105263158e-06,
|
| 78 |
+
"loss": 1.1641,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.02631578947368421,
|
| 83 |
+
"grad_norm": 1.9595885172581249,
|
| 84 |
+
"learning_rate": 5.263157894736842e-06,
|
| 85 |
+
"loss": 1.2251,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.02894736842105263,
|
| 90 |
+
"grad_norm": 1.9931664072286583,
|
| 91 |
+
"learning_rate": 5.789473684210527e-06,
|
| 92 |
+
"loss": 1.1715,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.031578947368421054,
|
| 97 |
+
"grad_norm": 2.1154232990627757,
|
| 98 |
+
"learning_rate": 6.31578947368421e-06,
|
| 99 |
+
"loss": 1.1263,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.034210526315789476,
|
| 104 |
+
"grad_norm": 1.712961421918351,
|
| 105 |
+
"learning_rate": 6.842105263157896e-06,
|
| 106 |
+
"loss": 1.1846,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03684210526315789,
|
| 111 |
+
"grad_norm": 1.907897309472424,
|
| 112 |
+
"learning_rate": 7.368421052631579e-06,
|
| 113 |
+
"loss": 1.1599,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.039473684210526314,
|
| 118 |
+
"grad_norm": 1.8618405884575073,
|
| 119 |
+
"learning_rate": 7.894736842105265e-06,
|
| 120 |
+
"loss": 1.1446,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.042105263157894736,
|
| 125 |
+
"grad_norm": 2.003611558479045,
|
| 126 |
+
"learning_rate": 8.421052631578948e-06,
|
| 127 |
+
"loss": 1.2148,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04473684210526316,
|
| 132 |
+
"grad_norm": 1.9509755028786422,
|
| 133 |
+
"learning_rate": 8.947368421052632e-06,
|
| 134 |
+
"loss": 1.2278,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.04736842105263158,
|
| 139 |
+
"grad_norm": 2.227455915272786,
|
| 140 |
+
"learning_rate": 9.473684210526315e-06,
|
| 141 |
+
"loss": 1.1945,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05,
|
| 146 |
+
"grad_norm": 2.2089132290988966,
|
| 147 |
+
"learning_rate": 1e-05,
|
| 148 |
+
"loss": 1.2186,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05263157894736842,
|
| 153 |
+
"grad_norm": 2.0403472473994992,
|
| 154 |
+
"learning_rate": 1.0526315789473684e-05,
|
| 155 |
+
"loss": 1.244,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05526315789473684,
|
| 160 |
+
"grad_norm": 2.07699667956044,
|
| 161 |
+
"learning_rate": 1.105263157894737e-05,
|
| 162 |
+
"loss": 1.2287,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05789473684210526,
|
| 167 |
+
"grad_norm": 1.8269346866586127,
|
| 168 |
+
"learning_rate": 1.1578947368421053e-05,
|
| 169 |
+
"loss": 1.2289,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.060526315789473685,
|
| 174 |
+
"grad_norm": 1.9416484545897956,
|
| 175 |
+
"learning_rate": 1.2105263157894737e-05,
|
| 176 |
+
"loss": 1.1656,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06315789473684211,
|
| 181 |
+
"grad_norm": 1.8329146478042504,
|
| 182 |
+
"learning_rate": 1.263157894736842e-05,
|
| 183 |
+
"loss": 1.2144,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06578947368421052,
|
| 188 |
+
"grad_norm": 1.9961976737672504,
|
| 189 |
+
"learning_rate": 1.3157894736842108e-05,
|
| 190 |
+
"loss": 1.1669,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06842105263157895,
|
| 195 |
+
"grad_norm": 1.817199311046236,
|
| 196 |
+
"learning_rate": 1.3684210526315791e-05,
|
| 197 |
+
"loss": 1.2309,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07105263157894737,
|
| 202 |
+
"grad_norm": 1.7397043184426004,
|
| 203 |
+
"learning_rate": 1.4210526315789475e-05,
|
| 204 |
+
"loss": 1.1937,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07368421052631578,
|
| 209 |
+
"grad_norm": 2.045890849321593,
|
| 210 |
+
"learning_rate": 1.4736842105263159e-05,
|
| 211 |
+
"loss": 1.2241,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07631578947368421,
|
| 216 |
+
"grad_norm": 1.9435204128532164,
|
| 217 |
+
"learning_rate": 1.5263157894736846e-05,
|
| 218 |
+
"loss": 1.208,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07894736842105263,
|
| 223 |
+
"grad_norm": 3.270850219919583,
|
| 224 |
+
"learning_rate": 1.578947368421053e-05,
|
| 225 |
+
"loss": 1.1838,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08157894736842106,
|
| 230 |
+
"grad_norm": 1.9095124068691054,
|
| 231 |
+
"learning_rate": 1.6315789473684213e-05,
|
| 232 |
+
"loss": 1.2043,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08421052631578947,
|
| 237 |
+
"grad_norm": 2.2387627201274025,
|
| 238 |
+
"learning_rate": 1.6842105263157896e-05,
|
| 239 |
+
"loss": 1.2013,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.0868421052631579,
|
| 244 |
+
"grad_norm": 1.9325778214196014,
|
| 245 |
+
"learning_rate": 1.736842105263158e-05,
|
| 246 |
+
"loss": 1.211,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.08947368421052632,
|
| 251 |
+
"grad_norm": 2.0833292593807107,
|
| 252 |
+
"learning_rate": 1.7894736842105264e-05,
|
| 253 |
+
"loss": 1.2453,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09210526315789473,
|
| 258 |
+
"grad_norm": 1.927275946654803,
|
| 259 |
+
"learning_rate": 1.8421052631578947e-05,
|
| 260 |
+
"loss": 1.2558,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09473684210526316,
|
| 265 |
+
"grad_norm": 1.7879806074211535,
|
| 266 |
+
"learning_rate": 1.894736842105263e-05,
|
| 267 |
+
"loss": 1.2603,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09736842105263158,
|
| 272 |
+
"grad_norm": 1.6245970203120101,
|
| 273 |
+
"learning_rate": 1.9473684210526318e-05,
|
| 274 |
+
"loss": 1.285,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1,
|
| 279 |
+
"grad_norm": 1.7551464083695452,
|
| 280 |
+
"learning_rate": 2e-05,
|
| 281 |
+
"loss": 1.2406,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10263157894736842,
|
| 286 |
+
"grad_norm": 1.8960908258050249,
|
| 287 |
+
"learning_rate": 1.9999578095183126e-05,
|
| 288 |
+
"loss": 1.2636,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10526315789473684,
|
| 293 |
+
"grad_norm": 1.8238289222983086,
|
| 294 |
+
"learning_rate": 1.999831241633323e-05,
|
| 295 |
+
"loss": 1.2385,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.10789473684210527,
|
| 300 |
+
"grad_norm": 1.7131092211626127,
|
| 301 |
+
"learning_rate": 1.9996203070249516e-05,
|
| 302 |
+
"loss": 1.2057,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.11052631578947368,
|
| 307 |
+
"grad_norm": 1.9814641064127423,
|
| 308 |
+
"learning_rate": 1.9993250234920638e-05,
|
| 309 |
+
"loss": 1.2105,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11315789473684211,
|
| 314 |
+
"grad_norm": 1.764032181247943,
|
| 315 |
+
"learning_rate": 1.998945415950969e-05,
|
| 316 |
+
"loss": 1.2082,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11578947368421053,
|
| 321 |
+
"grad_norm": 1.7110028028184467,
|
| 322 |
+
"learning_rate": 1.9984815164333163e-05,
|
| 323 |
+
"loss": 1.2292,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.11842105263157894,
|
| 328 |
+
"grad_norm": 1.9617301301812387,
|
| 329 |
+
"learning_rate": 1.9979333640833947e-05,
|
| 330 |
+
"loss": 1.2204,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12105263157894737,
|
| 335 |
+
"grad_norm": 1.7477888973288305,
|
| 336 |
+
"learning_rate": 1.9973010051548274e-05,
|
| 337 |
+
"loss": 1.249,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12368421052631579,
|
| 342 |
+
"grad_norm": 1.7276840169609364,
|
| 343 |
+
"learning_rate": 1.99658449300667e-05,
|
| 344 |
+
"loss": 1.1936,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12631578947368421,
|
| 349 |
+
"grad_norm": 1.858438572075226,
|
| 350 |
+
"learning_rate": 1.9957838880989076e-05,
|
| 351 |
+
"loss": 1.2914,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.12894736842105264,
|
| 356 |
+
"grad_norm": 1.6258008392447836,
|
| 357 |
+
"learning_rate": 1.9948992579873538e-05,
|
| 358 |
+
"loss": 1.2453,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13157894736842105,
|
| 363 |
+
"grad_norm": 1.6589841824832805,
|
| 364 |
+
"learning_rate": 1.9939306773179498e-05,
|
| 365 |
+
"loss": 1.2091,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13421052631578947,
|
| 370 |
+
"grad_norm": 1.6489987345538428,
|
| 371 |
+
"learning_rate": 1.992878227820465e-05,
|
| 372 |
+
"loss": 1.2223,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1368421052631579,
|
| 377 |
+
"grad_norm": 1.736814397762254,
|
| 378 |
+
"learning_rate": 1.9917419983016025e-05,
|
| 379 |
+
"loss": 1.2464,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.1394736842105263,
|
| 384 |
+
"grad_norm": 1.612061991991375,
|
| 385 |
+
"learning_rate": 1.990522084637503e-05,
|
| 386 |
+
"loss": 1.2334,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.14210526315789473,
|
| 391 |
+
"grad_norm": 1.634510448068168,
|
| 392 |
+
"learning_rate": 1.989218589765658e-05,
|
| 393 |
+
"loss": 1.2618,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14473684210526316,
|
| 398 |
+
"grad_norm": 1.7336110112785093,
|
| 399 |
+
"learning_rate": 1.9878316236762195e-05,
|
| 400 |
+
"loss": 1.2533,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14736842105263157,
|
| 405 |
+
"grad_norm": 1.8212242851158706,
|
| 406 |
+
"learning_rate": 1.9863613034027224e-05,
|
| 407 |
+
"loss": 1.2929,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.15,
|
| 412 |
+
"grad_norm": 1.7068273105310745,
|
| 413 |
+
"learning_rate": 1.9848077530122083e-05,
|
| 414 |
+
"loss": 1.1776,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.15263157894736842,
|
| 419 |
+
"grad_norm": 1.6973152439927286,
|
| 420 |
+
"learning_rate": 1.9831711035947552e-05,
|
| 421 |
+
"loss": 1.2079,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15526315789473685,
|
| 426 |
+
"grad_norm": 1.7272781391952396,
|
| 427 |
+
"learning_rate": 1.981451493252418e-05,
|
| 428 |
+
"loss": 1.2529,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.15789473684210525,
|
| 433 |
+
"grad_norm": 1.8086226050669576,
|
| 434 |
+
"learning_rate": 1.979649067087574e-05,
|
| 435 |
+
"loss": 1.2245,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.16052631578947368,
|
| 440 |
+
"grad_norm": 1.7065793845674917,
|
| 441 |
+
"learning_rate": 1.9777639771906795e-05,
|
| 442 |
+
"loss": 1.2693,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.1631578947368421,
|
| 447 |
+
"grad_norm": 1.5865552591333323,
|
| 448 |
+
"learning_rate": 1.9757963826274357e-05,
|
| 449 |
+
"loss": 1.2161,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16578947368421051,
|
| 454 |
+
"grad_norm": 1.5842481812204574,
|
| 455 |
+
"learning_rate": 1.973746449425368e-05,
|
| 456 |
+
"loss": 1.1984,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.16842105263157894,
|
| 461 |
+
"grad_norm": 1.6377526238069449,
|
| 462 |
+
"learning_rate": 1.971614350559814e-05,
|
| 463 |
+
"loss": 1.2238,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.17105263157894737,
|
| 468 |
+
"grad_norm": 1.4997127407445148,
|
| 469 |
+
"learning_rate": 1.9694002659393306e-05,
|
| 470 |
+
"loss": 1.2261,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.1736842105263158,
|
| 475 |
+
"grad_norm": 2.0809171217051583,
|
| 476 |
+
"learning_rate": 1.967104382390511e-05,
|
| 477 |
+
"loss": 1.2353,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.1763157894736842,
|
| 482 |
+
"grad_norm": 1.5420942514065212,
|
| 483 |
+
"learning_rate": 1.9647268936422204e-05,
|
| 484 |
+
"loss": 1.1808,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.17894736842105263,
|
| 489 |
+
"grad_norm": 1.7331848458071615,
|
| 490 |
+
"learning_rate": 1.9622680003092503e-05,
|
| 491 |
+
"loss": 1.2808,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18157894736842106,
|
| 496 |
+
"grad_norm": 1.642676114735854,
|
| 497 |
+
"learning_rate": 1.9597279098753893e-05,
|
| 498 |
+
"loss": 1.1869,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.18421052631578946,
|
| 503 |
+
"grad_norm": 1.7245853030106764,
|
| 504 |
+
"learning_rate": 1.9571068366759143e-05,
|
| 505 |
+
"loss": 1.272,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.1868421052631579,
|
| 510 |
+
"grad_norm": 1.611354187249786,
|
| 511 |
+
"learning_rate": 1.9544050018795076e-05,
|
| 512 |
+
"loss": 1.2521,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.18947368421052632,
|
| 517 |
+
"grad_norm": 1.604892439509072,
|
| 518 |
+
"learning_rate": 1.951622633469592e-05,
|
| 519 |
+
"loss": 1.2039,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19210526315789472,
|
| 524 |
+
"grad_norm": 1.5654642508374634,
|
| 525 |
+
"learning_rate": 1.9487599662250945e-05,
|
| 526 |
+
"loss": 1.2186,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.19473684210526315,
|
| 531 |
+
"grad_norm": 1.5636346341285678,
|
| 532 |
+
"learning_rate": 1.9458172417006347e-05,
|
| 533 |
+
"loss": 1.2436,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.19736842105263158,
|
| 538 |
+
"grad_norm": 1.7025910710749705,
|
| 539 |
+
"learning_rate": 1.9427947082061432e-05,
|
| 540 |
+
"loss": 1.2455,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.2,
|
| 545 |
+
"grad_norm": 1.7656433674820964,
|
| 546 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 547 |
+
"loss": 1.2548,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2026315789473684,
|
| 552 |
+
"grad_norm": 1.6243476505198133,
|
| 553 |
+
"learning_rate": 1.936511241197055e-05,
|
| 554 |
+
"loss": 1.3026,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.20526315789473684,
|
| 559 |
+
"grad_norm": 1.5869818361559598,
|
| 560 |
+
"learning_rate": 1.933250837887457e-05,
|
| 561 |
+
"loss": 1.2697,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.20789473684210527,
|
| 566 |
+
"grad_norm": 1.7074439932161543,
|
| 567 |
+
"learning_rate": 1.929911685973088e-05,
|
| 568 |
+
"loss": 1.2321,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.21052631578947367,
|
| 573 |
+
"grad_norm": 1.66975384315729,
|
| 574 |
+
"learning_rate": 1.9264940672148018e-05,
|
| 575 |
+
"loss": 1.1919,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2131578947368421,
|
| 580 |
+
"grad_norm": 1.5614279544765697,
|
| 581 |
+
"learning_rate": 1.922998269994563e-05,
|
| 582 |
+
"loss": 1.21,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.21578947368421053,
|
| 587 |
+
"grad_norm": 1.5869104530304803,
|
| 588 |
+
"learning_rate": 1.9194245892911077e-05,
|
| 589 |
+
"loss": 1.2005,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.21842105263157896,
|
| 594 |
+
"grad_norm": 1.5005378779293412,
|
| 595 |
+
"learning_rate": 1.9157733266550577e-05,
|
| 596 |
+
"loss": 1.2235,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.22105263157894736,
|
| 601 |
+
"grad_norm": 1.7626666578807375,
|
| 602 |
+
"learning_rate": 1.9120447901834708e-05,
|
| 603 |
+
"loss": 1.2067,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2236842105263158,
|
| 608 |
+
"grad_norm": 1.6204644035149893,
|
| 609 |
+
"learning_rate": 1.9082392944938467e-05,
|
| 610 |
+
"loss": 1.2733,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.22631578947368422,
|
| 615 |
+
"grad_norm": 1.630060745845602,
|
| 616 |
+
"learning_rate": 1.9043571606975776e-05,
|
| 617 |
+
"loss": 1.2548,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.22894736842105262,
|
| 622 |
+
"grad_norm": 1.5126387044839233,
|
| 623 |
+
"learning_rate": 1.9003987163728535e-05,
|
| 624 |
+
"loss": 1.2,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.23157894736842105,
|
| 629 |
+
"grad_norm": 1.6073465137634237,
|
| 630 |
+
"learning_rate": 1.8963642955370203e-05,
|
| 631 |
+
"loss": 1.2256,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23421052631578948,
|
| 636 |
+
"grad_norm": 1.697359488197219,
|
| 637 |
+
"learning_rate": 1.8922542386183942e-05,
|
| 638 |
+
"loss": 1.2801,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.23684210526315788,
|
| 643 |
+
"grad_norm": 1.4698043032405201,
|
| 644 |
+
"learning_rate": 1.888068892427538e-05,
|
| 645 |
+
"loss": 1.1803,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2394736842105263,
|
| 650 |
+
"grad_norm": 1.505416525863708,
|
| 651 |
+
"learning_rate": 1.8838086101279946e-05,
|
| 652 |
+
"loss": 1.2403,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24210526315789474,
|
| 657 |
+
"grad_norm": 1.7591893486363919,
|
| 658 |
+
"learning_rate": 1.879473751206489e-05,
|
| 659 |
+
"loss": 1.23,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.24473684210526317,
|
| 664 |
+
"grad_norm": 1.6372396258203044,
|
| 665 |
+
"learning_rate": 1.875064681442594e-05,
|
| 666 |
+
"loss": 1.2144,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.24736842105263157,
|
| 671 |
+
"grad_norm": 1.5964554118260599,
|
| 672 |
+
"learning_rate": 1.8705817728778626e-05,
|
| 673 |
+
"loss": 1.2389,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.25,
|
| 678 |
+
"grad_norm": 1.5889929908257716,
|
| 679 |
+
"learning_rate": 1.866025403784439e-05,
|
| 680 |
+
"loss": 1.1556,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.25263157894736843,
|
| 685 |
+
"grad_norm": 1.4618065158181492,
|
| 686 |
+
"learning_rate": 1.8613959586331364e-05,
|
| 687 |
+
"loss": 1.2511,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.25526315789473686,
|
| 692 |
+
"grad_norm": 1.5332613540370528,
|
| 693 |
+
"learning_rate": 1.8566938280609965e-05,
|
| 694 |
+
"loss": 1.2578,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2578947368421053,
|
| 699 |
+
"grad_norm": 1.5645156940351497,
|
| 700 |
+
"learning_rate": 1.851919408838327e-05,
|
| 701 |
+
"loss": 1.2014,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.26052631578947366,
|
| 706 |
+
"grad_norm": 1.6458424744447175,
|
| 707 |
+
"learning_rate": 1.847073103835222e-05,
|
| 708 |
+
"loss": 1.251,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2631578947368421,
|
| 713 |
+
"grad_norm": 1.6298298976442505,
|
| 714 |
+
"learning_rate": 1.842155321987566e-05,
|
| 715 |
+
"loss": 1.237,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.2657894736842105,
|
| 720 |
+
"grad_norm": 1.4753056372696693,
|
| 721 |
+
"learning_rate": 1.8371664782625287e-05,
|
| 722 |
+
"loss": 1.2234,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.26842105263157895,
|
| 727 |
+
"grad_norm": 1.5302926579741902,
|
| 728 |
+
"learning_rate": 1.8321069936235503e-05,
|
| 729 |
+
"loss": 1.2159,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2710526315789474,
|
| 734 |
+
"grad_norm": 1.4566906526101497,
|
| 735 |
+
"learning_rate": 1.8269772949948185e-05,
|
| 736 |
+
"loss": 1.2219,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.2736842105263158,
|
| 741 |
+
"grad_norm": 1.4798388569320753,
|
| 742 |
+
"learning_rate": 1.821777815225245e-05,
|
| 743 |
+
"loss": 1.2168,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.27631578947368424,
|
| 748 |
+
"grad_norm": 1.578257631631102,
|
| 749 |
+
"learning_rate": 1.816508993051943e-05,
|
| 750 |
+
"loss": 1.2145,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2789473684210526,
|
| 755 |
+
"grad_norm": 1.535918888434136,
|
| 756 |
+
"learning_rate": 1.8111712730632024e-05,
|
| 757 |
+
"loss": 1.2309,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.28157894736842104,
|
| 762 |
+
"grad_norm": 1.56754973743785,
|
| 763 |
+
"learning_rate": 1.8057651056609784e-05,
|
| 764 |
+
"loss": 1.2554,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.28421052631578947,
|
| 769 |
+
"grad_norm": 1.5522488842058326,
|
| 770 |
+
"learning_rate": 1.800290947022884e-05,
|
| 771 |
+
"loss": 1.2529,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2868421052631579,
|
| 776 |
+
"grad_norm": 1.475576507279166,
|
| 777 |
+
"learning_rate": 1.7947492590636998e-05,
|
| 778 |
+
"loss": 1.268,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.2894736842105263,
|
| 783 |
+
"grad_norm": 1.5056156987526548,
|
| 784 |
+
"learning_rate": 1.789140509396394e-05,
|
| 785 |
+
"loss": 1.2427,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.29210526315789476,
|
| 790 |
+
"grad_norm": 1.643462413142722,
|
| 791 |
+
"learning_rate": 1.7834651712926662e-05,
|
| 792 |
+
"loss": 1.2299,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.29473684210526313,
|
| 797 |
+
"grad_norm": 1.636081381079421,
|
| 798 |
+
"learning_rate": 1.777723723643014e-05,
|
| 799 |
+
"loss": 1.2792,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.29736842105263156,
|
| 804 |
+
"grad_norm": 1.6196644995812366,
|
| 805 |
+
"learning_rate": 1.771916650916321e-05,
|
| 806 |
+
"loss": 1.222,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3,
|
| 811 |
+
"grad_norm": 1.4763314093013102,
|
| 812 |
+
"learning_rate": 1.766044443118978e-05,
|
| 813 |
+
"loss": 1.2511,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3026315789473684,
|
| 818 |
+
"grad_norm": 1.555093791715219,
|
| 819 |
+
"learning_rate": 1.7601075957535366e-05,
|
| 820 |
+
"loss": 1.205,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.30526315789473685,
|
| 825 |
+
"grad_norm": 1.524333754681362,
|
| 826 |
+
"learning_rate": 1.7541066097768965e-05,
|
| 827 |
+
"loss": 1.2326,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3078947368421053,
|
| 832 |
+
"grad_norm": 1.4619146723611067,
|
| 833 |
+
"learning_rate": 1.7480419915580357e-05,
|
| 834 |
+
"loss": 1.2466,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3105263157894737,
|
| 839 |
+
"grad_norm": 1.4785833344608155,
|
| 840 |
+
"learning_rate": 1.7419142528352815e-05,
|
| 841 |
+
"loss": 1.2773,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3131578947368421,
|
| 846 |
+
"grad_norm": 1.4743114052842876,
|
| 847 |
+
"learning_rate": 1.735723910673132e-05,
|
| 848 |
+
"loss": 1.224,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3157894736842105,
|
| 853 |
+
"grad_norm": 1.6746586071080014,
|
| 854 |
+
"learning_rate": 1.729471487418621e-05,
|
| 855 |
+
"loss": 1.2031,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.31842105263157894,
|
| 860 |
+
"grad_norm": 11.560581182062958,
|
| 861 |
+
"learning_rate": 1.723157510657247e-05,
|
| 862 |
+
"loss": 1.2641,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.32105263157894737,
|
| 867 |
+
"grad_norm": 1.5785699977927574,
|
| 868 |
+
"learning_rate": 1.7167825131684516e-05,
|
| 869 |
+
"loss": 1.2406,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3236842105263158,
|
| 874 |
+
"grad_norm": 1.4907215745859081,
|
| 875 |
+
"learning_rate": 1.710347032880664e-05,
|
| 876 |
+
"loss": 1.2393,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.3263157894736842,
|
| 881 |
+
"grad_norm": 1.4873798233212445,
|
| 882 |
+
"learning_rate": 1.7038516128259118e-05,
|
| 883 |
+
"loss": 1.2613,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.32894736842105265,
|
| 888 |
+
"grad_norm": 1.5996710317262621,
|
| 889 |
+
"learning_rate": 1.6972968010939953e-05,
|
| 890 |
+
"loss": 1.226,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.33157894736842103,
|
| 895 |
+
"grad_norm": 1.5258666469896836,
|
| 896 |
+
"learning_rate": 1.6906831507862446e-05,
|
| 897 |
+
"loss": 1.1765,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.33421052631578946,
|
| 902 |
+
"grad_norm": 1.65626890509294,
|
| 903 |
+
"learning_rate": 1.6840112199688432e-05,
|
| 904 |
+
"loss": 1.2042,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3368421052631579,
|
| 909 |
+
"grad_norm": 1.5999899196235836,
|
| 910 |
+
"learning_rate": 1.6772815716257414e-05,
|
| 911 |
+
"loss": 1.2165,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3394736842105263,
|
| 916 |
+
"grad_norm": 1.539114071232729,
|
| 917 |
+
"learning_rate": 1.670494773611149e-05,
|
| 918 |
+
"loss": 1.2325,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.34210526315789475,
|
| 923 |
+
"grad_norm": 1.5498968903780033,
|
| 924 |
+
"learning_rate": 1.6636513986016215e-05,
|
| 925 |
+
"loss": 1.2489,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3447368421052632,
|
| 930 |
+
"grad_norm": 1.6929154737187613,
|
| 931 |
+
"learning_rate": 1.6567520240477344e-05,
|
| 932 |
+
"loss": 1.2733,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3473684210526316,
|
| 937 |
+
"grad_norm": 1.5186251712267134,
|
| 938 |
+
"learning_rate": 1.64979723212536e-05,
|
| 939 |
+
"loss": 1.2163,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.35,
|
| 944 |
+
"grad_norm": 1.5656728169955136,
|
| 945 |
+
"learning_rate": 1.6427876096865394e-05,
|
| 946 |
+
"loss": 1.1686,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3526315789473684,
|
| 951 |
+
"grad_norm": 1.7024084700143047,
|
| 952 |
+
"learning_rate": 1.6357237482099682e-05,
|
| 953 |
+
"loss": 1.2607,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.35526315789473684,
|
| 958 |
+
"grad_norm": 1.5453936941229016,
|
| 959 |
+
"learning_rate": 1.6286062437510823e-05,
|
| 960 |
+
"loss": 1.2788,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.35789473684210527,
|
| 965 |
+
"grad_norm": 1.5519495902650802,
|
| 966 |
+
"learning_rate": 1.621435696891765e-05,
|
| 967 |
+
"loss": 1.2323,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.3605263157894737,
|
| 972 |
+
"grad_norm": 1.4847378374853688,
|
| 973 |
+
"learning_rate": 1.6142127126896682e-05,
|
| 974 |
+
"loss": 1.2871,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.3631578947368421,
|
| 979 |
+
"grad_norm": 1.426933507759982,
|
| 980 |
+
"learning_rate": 1.606937900627157e-05,
|
| 981 |
+
"loss": 1.1768,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.36578947368421055,
|
| 986 |
+
"grad_norm": 1.4919758909790284,
|
| 987 |
+
"learning_rate": 1.5996118745598817e-05,
|
| 988 |
+
"loss": 1.2079,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3684210526315789,
|
| 993 |
+
"grad_norm": 1.4485312135827444,
|
| 994 |
+
"learning_rate": 1.5922352526649803e-05,
|
| 995 |
+
"loss": 1.2442,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.37105263157894736,
|
| 1000 |
+
"grad_norm": 1.4755986147337776,
|
| 1001 |
+
"learning_rate": 1.584808657388914e-05,
|
| 1002 |
+
"loss": 1.2329,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3736842105263158,
|
| 1007 |
+
"grad_norm": 1.5552070636270465,
|
| 1008 |
+
"learning_rate": 1.5773327153949465e-05,
|
| 1009 |
+
"loss": 1.2234,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.3763157894736842,
|
| 1014 |
+
"grad_norm": 1.9398418581511307,
|
| 1015 |
+
"learning_rate": 1.5698080575102662e-05,
|
| 1016 |
+
"loss": 1.2528,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.37894736842105264,
|
| 1021 |
+
"grad_norm": 1.5594003459102397,
|
| 1022 |
+
"learning_rate": 1.5622353186727542e-05,
|
| 1023 |
+
"loss": 1.2619,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3815789473684211,
|
| 1028 |
+
"grad_norm": 1.5755383497580695,
|
| 1029 |
+
"learning_rate": 1.554615137877409e-05,
|
| 1030 |
+
"loss": 1.2643,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.38421052631578945,
|
| 1035 |
+
"grad_norm": 1.5535461809815971,
|
| 1036 |
+
"learning_rate": 1.5469481581224274e-05,
|
| 1037 |
+
"loss": 1.1972,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.3868421052631579,
|
| 1042 |
+
"grad_norm": 1.5115819913848945,
|
| 1043 |
+
"learning_rate": 1.5392350263549462e-05,
|
| 1044 |
+
"loss": 1.243,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.3894736842105263,
|
| 1049 |
+
"grad_norm": 1.527666218311888,
|
| 1050 |
+
"learning_rate": 1.531476393416456e-05,
|
| 1051 |
+
"loss": 1.2062,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.39210526315789473,
|
| 1056 |
+
"grad_norm": 1.4753990139855222,
|
| 1057 |
+
"learning_rate": 1.523672913987878e-05,
|
| 1058 |
+
"loss": 1.218,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.39473684210526316,
|
| 1063 |
+
"grad_norm": 1.4294571826626545,
|
| 1064 |
+
"learning_rate": 1.5158252465343242e-05,
|
| 1065 |
+
"loss": 1.1627,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.3973684210526316,
|
| 1070 |
+
"grad_norm": 1.5436073392476704,
|
| 1071 |
+
"learning_rate": 1.5079340532495344e-05,
|
| 1072 |
+
"loss": 1.1995,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.4,
|
| 1077 |
+
"grad_norm": 1.498938650619184,
|
| 1078 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 1079 |
+
"loss": 1.2387,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4026315789473684,
|
| 1084 |
+
"grad_norm": 1.4966516858258112,
|
| 1085 |
+
"learning_rate": 1.4920237562687784e-05,
|
| 1086 |
+
"loss": 1.2343,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4052631578947368,
|
| 1091 |
+
"grad_norm": 1.600369428001101,
|
| 1092 |
+
"learning_rate": 1.4840059950989992e-05,
|
| 1093 |
+
"loss": 1.2335,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.40789473684210525,
|
| 1098 |
+
"grad_norm": 1.6146064486984097,
|
| 1099 |
+
"learning_rate": 1.4759473930370738e-05,
|
| 1100 |
+
"loss": 1.2445,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4105263157894737,
|
| 1105 |
+
"grad_norm": 1.4179121925460856,
|
| 1106 |
+
"learning_rate": 1.467848630075608e-05,
|
| 1107 |
+
"loss": 1.2005,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4131578947368421,
|
| 1112 |
+
"grad_norm": 1.4026365243596637,
|
| 1113 |
+
"learning_rate": 1.4597103895960228e-05,
|
| 1114 |
+
"loss": 1.1707,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.41578947368421054,
|
| 1119 |
+
"grad_norm": 1.608224849561234,
|
| 1120 |
+
"learning_rate": 1.4515333583108896e-05,
|
| 1121 |
+
"loss": 1.2172,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.41842105263157897,
|
| 1126 |
+
"grad_norm": 1.5892555814004448,
|
| 1127 |
+
"learning_rate": 1.4433182262059861e-05,
|
| 1128 |
+
"loss": 1.2437,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.42105263157894735,
|
| 1133 |
+
"grad_norm": 1.5528172119361345,
|
| 1134 |
+
"learning_rate": 1.4350656864820733e-05,
|
| 1135 |
+
"loss": 1.2458,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4236842105263158,
|
| 1140 |
+
"grad_norm": 1.4895924962034135,
|
| 1141 |
+
"learning_rate": 1.4267764354964038e-05,
|
| 1142 |
+
"loss": 1.217,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4263157894736842,
|
| 1147 |
+
"grad_norm": 1.4853890475688154,
|
| 1148 |
+
"learning_rate": 1.4184511727039612e-05,
|
| 1149 |
+
"loss": 1.1886,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.42894736842105263,
|
| 1154 |
+
"grad_norm": 1.484648338587035,
|
| 1155 |
+
"learning_rate": 1.4100906005984404e-05,
|
| 1156 |
+
"loss": 1.1946,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.43157894736842106,
|
| 1161 |
+
"grad_norm": 1.4930103966077508,
|
| 1162 |
+
"learning_rate": 1.4016954246529697e-05,
|
| 1163 |
+
"loss": 1.168,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4342105263157895,
|
| 1168 |
+
"grad_norm": 1.5631231114882895,
|
| 1169 |
+
"learning_rate": 1.3932663532605832e-05,
|
| 1170 |
+
"loss": 1.1962,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4368421052631579,
|
| 1175 |
+
"grad_norm": 1.4246463220095698,
|
| 1176 |
+
"learning_rate": 1.3848040976744459e-05,
|
| 1177 |
+
"loss": 1.1903,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.4394736842105263,
|
| 1182 |
+
"grad_norm": 1.652743816078508,
|
| 1183 |
+
"learning_rate": 1.3763093719478357e-05,
|
| 1184 |
+
"loss": 1.1791,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.4421052631578947,
|
| 1189 |
+
"grad_norm": 1.5195763177089594,
|
| 1190 |
+
"learning_rate": 1.3677828928738934e-05,
|
| 1191 |
+
"loss": 1.2274,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.44473684210526315,
|
| 1196 |
+
"grad_norm": 1.418736373104399,
|
| 1197 |
+
"learning_rate": 1.3592253799251377e-05,
|
| 1198 |
+
"loss": 1.203,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.4473684210526316,
|
| 1203 |
+
"grad_norm": 1.4260404389599979,
|
| 1204 |
+
"learning_rate": 1.3506375551927546e-05,
|
| 1205 |
+
"loss": 1.1997,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.45,
|
| 1210 |
+
"grad_norm": 1.6460529001511097,
|
| 1211 |
+
"learning_rate": 1.342020143325669e-05,
|
| 1212 |
+
"loss": 1.237,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.45263157894736844,
|
| 1217 |
+
"grad_norm": 1.486629400358489,
|
| 1218 |
+
"learning_rate": 1.3333738714693958e-05,
|
| 1219 |
+
"loss": 1.247,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.45526315789473687,
|
| 1224 |
+
"grad_norm": 1.3972348468520703,
|
| 1225 |
+
"learning_rate": 1.3246994692046837e-05,
|
| 1226 |
+
"loss": 1.2234,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.45789473684210524,
|
| 1231 |
+
"grad_norm": 1.6487432488778413,
|
| 1232 |
+
"learning_rate": 1.3159976684859528e-05,
|
| 1233 |
+
"loss": 1.2329,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.4605263157894737,
|
| 1238 |
+
"grad_norm": 1.3823799200592426,
|
| 1239 |
+
"learning_rate": 1.3072692035795305e-05,
|
| 1240 |
+
"loss": 1.1959,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4631578947368421,
|
| 1245 |
+
"grad_norm": 1.4633369850606546,
|
| 1246 |
+
"learning_rate": 1.2985148110016947e-05,
|
| 1247 |
+
"loss": 1.1918,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.46578947368421053,
|
| 1252 |
+
"grad_norm": 1.4284040647353275,
|
| 1253 |
+
"learning_rate": 1.2897352294565248e-05,
|
| 1254 |
+
"loss": 1.206,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.46842105263157896,
|
| 1259 |
+
"grad_norm": 1.557769882746258,
|
| 1260 |
+
"learning_rate": 1.2809311997735697e-05,
|
| 1261 |
+
"loss": 1.1774,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.4710526315789474,
|
| 1266 |
+
"grad_norm": 1.696104251000462,
|
| 1267 |
+
"learning_rate": 1.2721034648453353e-05,
|
| 1268 |
+
"loss": 1.2192,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.47368421052631576,
|
| 1273 |
+
"grad_norm": 1.3992290817350297,
|
| 1274 |
+
"learning_rate": 1.2632527695645993e-05,
|
| 1275 |
+
"loss": 1.1794,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4763157894736842,
|
| 1280 |
+
"grad_norm": 1.526786732330722,
|
| 1281 |
+
"learning_rate": 1.2543798607615566e-05,
|
| 1282 |
+
"loss": 1.1735,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.4789473684210526,
|
| 1287 |
+
"grad_norm": 1.4303793807251683,
|
| 1288 |
+
"learning_rate": 1.2454854871407993e-05,
|
| 1289 |
+
"loss": 1.1869,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.48157894736842105,
|
| 1294 |
+
"grad_norm": 1.4381387780322723,
|
| 1295 |
+
"learning_rate": 1.2365703992181425e-05,
|
| 1296 |
+
"loss": 1.1637,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.4842105263157895,
|
| 1301 |
+
"grad_norm": 1.347581682014551,
|
| 1302 |
+
"learning_rate": 1.2276353492572937e-05,
|
| 1303 |
+
"loss": 1.2333,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.4868421052631579,
|
| 1308 |
+
"grad_norm": 1.4212086902989138,
|
| 1309 |
+
"learning_rate": 1.218681091206376e-05,
|
| 1310 |
+
"loss": 1.2057,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.48947368421052634,
|
| 1315 |
+
"grad_norm": 1.4767598997567368,
|
| 1316 |
+
"learning_rate": 1.2097083806343104e-05,
|
| 1317 |
+
"loss": 1.2078,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.4921052631578947,
|
| 1322 |
+
"grad_norm": 1.6766286206430314,
|
| 1323 |
+
"learning_rate": 1.2007179746670592e-05,
|
| 1324 |
+
"loss": 1.178,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.49473684210526314,
|
| 1329 |
+
"grad_norm": 1.3676201703331183,
|
| 1330 |
+
"learning_rate": 1.1917106319237386e-05,
|
| 1331 |
+
"loss": 1.1578,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.49736842105263157,
|
| 1336 |
+
"grad_norm": 2.0706388440208663,
|
| 1337 |
+
"learning_rate": 1.1826871124526072e-05,
|
| 1338 |
+
"loss": 1.196,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5,
|
| 1343 |
+
"grad_norm": 1.507077620161342,
|
| 1344 |
+
"learning_rate": 1.1736481776669307e-05,
|
| 1345 |
+
"loss": 1.2277,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5026315789473684,
|
| 1350 |
+
"grad_norm": 1.4969439156513775,
|
| 1351 |
+
"learning_rate": 1.164594590280734e-05,
|
| 1352 |
+
"loss": 1.1897,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5052631578947369,
|
| 1357 |
+
"grad_norm": 1.413902082668003,
|
| 1358 |
+
"learning_rate": 1.1555271142444433e-05,
|
| 1359 |
+
"loss": 1.1878,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5078947368421053,
|
| 1364 |
+
"grad_norm": 1.404071675870193,
|
| 1365 |
+
"learning_rate": 1.1464465146804218e-05,
|
| 1366 |
+
"loss": 1.2204,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5105263157894737,
|
| 1371 |
+
"grad_norm": 1.420066039178241,
|
| 1372 |
+
"learning_rate": 1.1373535578184083e-05,
|
| 1373 |
+
"loss": 1.143,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5131578947368421,
|
| 1378 |
+
"grad_norm": 1.8886674026004764,
|
| 1379 |
+
"learning_rate": 1.1282490109308633e-05,
|
| 1380 |
+
"loss": 1.2461,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5157894736842106,
|
| 1385 |
+
"grad_norm": 1.4506368194815937,
|
| 1386 |
+
"learning_rate": 1.1191336422682237e-05,
|
| 1387 |
+
"loss": 1.2316,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5184210526315789,
|
| 1392 |
+
"grad_norm": 1.4848545614866153,
|
| 1393 |
+
"learning_rate": 1.1100082209940795e-05,
|
| 1394 |
+
"loss": 1.2514,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5210526315789473,
|
| 1399 |
+
"grad_norm": 1.3996822789360135,
|
| 1400 |
+
"learning_rate": 1.1008735171202685e-05,
|
| 1401 |
+
"loss": 1.2019,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5236842105263158,
|
| 1406 |
+
"grad_norm": 1.4330869323971005,
|
| 1407 |
+
"learning_rate": 1.0917303014419036e-05,
|
| 1408 |
+
"loss": 1.2031,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5263157894736842,
|
| 1413 |
+
"grad_norm": 1.4178423978580275,
|
| 1414 |
+
"learning_rate": 1.0825793454723325e-05,
|
| 1415 |
+
"loss": 1.266,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5289473684210526,
|
| 1420 |
+
"grad_norm": 1.5463727949510788,
|
| 1421 |
+
"learning_rate": 1.0734214213780355e-05,
|
| 1422 |
+
"loss": 1.1943,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.531578947368421,
|
| 1427 |
+
"grad_norm": 1.6232414845928462,
|
| 1428 |
+
"learning_rate": 1.0642573019134703e-05,
|
| 1429 |
+
"loss": 1.1882,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5342105263157895,
|
| 1434 |
+
"grad_norm": 1.3956188498038653,
|
| 1435 |
+
"learning_rate": 1.0550877603558656e-05,
|
| 1436 |
+
"loss": 1.2117,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5368421052631579,
|
| 1441 |
+
"grad_norm": 1.570355181053573,
|
| 1442 |
+
"learning_rate": 1.045913570439972e-05,
|
| 1443 |
+
"loss": 1.1844,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5394736842105263,
|
| 1448 |
+
"grad_norm": 1.4584214592171902,
|
| 1449 |
+
"learning_rate": 1.0367355062927726e-05,
|
| 1450 |
+
"loss": 1.1617,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5421052631578948,
|
| 1455 |
+
"grad_norm": 1.4365606902754935,
|
| 1456 |
+
"learning_rate": 1.0275543423681622e-05,
|
| 1457 |
+
"loss": 1.2129,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5447368421052632,
|
| 1462 |
+
"grad_norm": 1.5290557760088286,
|
| 1463 |
+
"learning_rate": 1.0183708533815975e-05,
|
| 1464 |
+
"loss": 1.2482,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5473684210526316,
|
| 1469 |
+
"grad_norm": 1.6278464681436842,
|
| 1470 |
+
"learning_rate": 1.0091858142447266e-05,
|
| 1471 |
+
"loss": 1.1489,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.55,
|
| 1476 |
+
"grad_norm": 1.5688826364703785,
|
| 1477 |
+
"learning_rate": 1e-05,
|
| 1478 |
+
"loss": 1.2095,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5526315789473685,
|
| 1483 |
+
"grad_norm": 1.397745444389566,
|
| 1484 |
+
"learning_rate": 9.908141857552737e-06,
|
| 1485 |
+
"loss": 1.2355,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5552631578947368,
|
| 1490 |
+
"grad_norm": 1.4583293118359388,
|
| 1491 |
+
"learning_rate": 9.816291466184025e-06,
|
| 1492 |
+
"loss": 1.1758,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5578947368421052,
|
| 1497 |
+
"grad_norm": 1.4603630721390348,
|
| 1498 |
+
"learning_rate": 9.724456576318383e-06,
|
| 1499 |
+
"loss": 1.1652,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5605263157894737,
|
| 1504 |
+
"grad_norm": 1.4549523975843028,
|
| 1505 |
+
"learning_rate": 9.632644937072277e-06,
|
| 1506 |
+
"loss": 1.1867,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5631578947368421,
|
| 1511 |
+
"grad_norm": 1.3729856367743716,
|
| 1512 |
+
"learning_rate": 9.540864295600282e-06,
|
| 1513 |
+
"loss": 1.2406,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5657894736842105,
|
| 1518 |
+
"grad_norm": 1.4783414676807372,
|
| 1519 |
+
"learning_rate": 9.449122396441344e-06,
|
| 1520 |
+
"loss": 1.1791,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5684210526315789,
|
| 1525 |
+
"grad_norm": 1.3872265988119108,
|
| 1526 |
+
"learning_rate": 9.3574269808653e-06,
|
| 1527 |
+
"loss": 1.2156,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5710526315789474,
|
| 1532 |
+
"grad_norm": 1.5787318597241866,
|
| 1533 |
+
"learning_rate": 9.265785786219647e-06,
|
| 1534 |
+
"loss": 1.2132,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5736842105263158,
|
| 1539 |
+
"grad_norm": 1.3724758535484003,
|
| 1540 |
+
"learning_rate": 9.174206545276678e-06,
|
| 1541 |
+
"loss": 1.1334,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5763157894736842,
|
| 1546 |
+
"grad_norm": 1.4868757284973377,
|
| 1547 |
+
"learning_rate": 9.082696985580964e-06,
|
| 1548 |
+
"loss": 1.2583,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5789473684210527,
|
| 1553 |
+
"grad_norm": 1.4676893755358382,
|
| 1554 |
+
"learning_rate": 8.991264828797319e-06,
|
| 1555 |
+
"loss": 1.2082,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5815789473684211,
|
| 1560 |
+
"grad_norm": 1.4127501220287753,
|
| 1561 |
+
"learning_rate": 8.899917790059208e-06,
|
| 1562 |
+
"loss": 1.211,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5842105263157895,
|
| 1567 |
+
"grad_norm": 1.332755695247283,
|
| 1568 |
+
"learning_rate": 8.808663577317765e-06,
|
| 1569 |
+
"loss": 1.1677,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5868421052631579,
|
| 1574 |
+
"grad_norm": 1.395357505892731,
|
| 1575 |
+
"learning_rate": 8.717509890691369e-06,
|
| 1576 |
+
"loss": 1.2085,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5894736842105263,
|
| 1581 |
+
"grad_norm": 1.6316377622079636,
|
| 1582 |
+
"learning_rate": 8.626464421815919e-06,
|
| 1583 |
+
"loss": 1.2065,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5921052631578947,
|
| 1588 |
+
"grad_norm": 1.4083070071992012,
|
| 1589 |
+
"learning_rate": 8.535534853195786e-06,
|
| 1590 |
+
"loss": 1.1612,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.5947368421052631,
|
| 1595 |
+
"grad_norm": 1.3425609671566616,
|
| 1596 |
+
"learning_rate": 8.444728857555572e-06,
|
| 1597 |
+
"loss": 1.2094,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.5973684210526315,
|
| 1602 |
+
"grad_norm": 1.5328975791749664,
|
| 1603 |
+
"learning_rate": 8.35405409719266e-06,
|
| 1604 |
+
"loss": 1.2155,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6,
|
| 1609 |
+
"grad_norm": 1.5590632136999736,
|
| 1610 |
+
"learning_rate": 8.263518223330698e-06,
|
| 1611 |
+
"loss": 1.1807,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6026315789473684,
|
| 1616 |
+
"grad_norm": 1.5215433696265592,
|
| 1617 |
+
"learning_rate": 8.173128875473933e-06,
|
| 1618 |
+
"loss": 1.183,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6052631578947368,
|
| 1623 |
+
"grad_norm": 1.6575473417500939,
|
| 1624 |
+
"learning_rate": 8.082893680762619e-06,
|
| 1625 |
+
"loss": 1.1353,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6078947368421053,
|
| 1630 |
+
"grad_norm": 1.460845514236017,
|
| 1631 |
+
"learning_rate": 7.99282025332941e-06,
|
| 1632 |
+
"loss": 1.219,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6105263157894737,
|
| 1637 |
+
"grad_norm": 1.4579638437828222,
|
| 1638 |
+
"learning_rate": 7.902916193656898e-06,
|
| 1639 |
+
"loss": 1.1703,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6131578947368421,
|
| 1644 |
+
"grad_norm": 1.4409394463378045,
|
| 1645 |
+
"learning_rate": 7.813189087936243e-06,
|
| 1646 |
+
"loss": 1.1735,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6157894736842106,
|
| 1651 |
+
"grad_norm": 1.5164224939867756,
|
| 1652 |
+
"learning_rate": 7.72364650742707e-06,
|
| 1653 |
+
"loss": 1.2259,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.618421052631579,
|
| 1658 |
+
"grad_norm": 1.4456788889526122,
|
| 1659 |
+
"learning_rate": 7.634296007818576e-06,
|
| 1660 |
+
"loss": 1.1755,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6210526315789474,
|
| 1665 |
+
"grad_norm": 1.5021862081854962,
|
| 1666 |
+
"learning_rate": 7.545145128592009e-06,
|
| 1667 |
+
"loss": 1.1817,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6236842105263158,
|
| 1672 |
+
"grad_norm": 1.3518521758976183,
|
| 1673 |
+
"learning_rate": 7.456201392384437e-06,
|
| 1674 |
+
"loss": 1.1728,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6263157894736842,
|
| 1679 |
+
"grad_norm": 1.4319948450891455,
|
| 1680 |
+
"learning_rate": 7.367472304354011e-06,
|
| 1681 |
+
"loss": 1.1546,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6289473684210526,
|
| 1686 |
+
"grad_norm": 1.5047899843401489,
|
| 1687 |
+
"learning_rate": 7.278965351546648e-06,
|
| 1688 |
+
"loss": 1.1986,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.631578947368421,
|
| 1693 |
+
"grad_norm": 1.4636280005741331,
|
| 1694 |
+
"learning_rate": 7.190688002264308e-06,
|
| 1695 |
+
"loss": 1.1226,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6342105263157894,
|
| 1700 |
+
"grad_norm": 1.4404270288293848,
|
| 1701 |
+
"learning_rate": 7.102647705434755e-06,
|
| 1702 |
+
"loss": 1.1784,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6368421052631579,
|
| 1707 |
+
"grad_norm": 1.6545110847593845,
|
| 1708 |
+
"learning_rate": 7.014851889983058e-06,
|
| 1709 |
+
"loss": 1.1839,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6394736842105263,
|
| 1714 |
+
"grad_norm": 1.4871827887195601,
|
| 1715 |
+
"learning_rate": 6.927307964204695e-06,
|
| 1716 |
+
"loss": 1.1887,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6421052631578947,
|
| 1721 |
+
"grad_norm": 1.576746407677905,
|
| 1722 |
+
"learning_rate": 6.840023315140476e-06,
|
| 1723 |
+
"loss": 1.1883,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6447368421052632,
|
| 1728 |
+
"grad_norm": 1.487715101079606,
|
| 1729 |
+
"learning_rate": 6.7530053079531664e-06,
|
| 1730 |
+
"loss": 1.1416,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6473684210526316,
|
| 1735 |
+
"grad_norm": 1.3965004840051898,
|
| 1736 |
+
"learning_rate": 6.666261285306048e-06,
|
| 1737 |
+
"loss": 1.1879,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.65,
|
| 1742 |
+
"grad_norm": 1.3874007274101692,
|
| 1743 |
+
"learning_rate": 6.579798566743314e-06,
|
| 1744 |
+
"loss": 1.145,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6526315789473685,
|
| 1749 |
+
"grad_norm": 1.5069537665388066,
|
| 1750 |
+
"learning_rate": 6.4936244480724575e-06,
|
| 1751 |
+
"loss": 1.214,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6552631578947369,
|
| 1756 |
+
"grad_norm": 1.335555512816458,
|
| 1757 |
+
"learning_rate": 6.407746200748628e-06,
|
| 1758 |
+
"loss": 1.2036,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6578947368421053,
|
| 1763 |
+
"grad_norm": 1.5356423714475045,
|
| 1764 |
+
"learning_rate": 6.322171071261071e-06,
|
| 1765 |
+
"loss": 1.2023,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6605263157894737,
|
| 1770 |
+
"grad_norm": 1.366112491986329,
|
| 1771 |
+
"learning_rate": 6.236906280521646e-06,
|
| 1772 |
+
"loss": 1.154,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6631578947368421,
|
| 1777 |
+
"grad_norm": 1.4961202079264277,
|
| 1778 |
+
"learning_rate": 6.151959023255545e-06,
|
| 1779 |
+
"loss": 1.1754,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6657894736842105,
|
| 1784 |
+
"grad_norm": 1.4946644203504424,
|
| 1785 |
+
"learning_rate": 6.067336467394169e-06,
|
| 1786 |
+
"loss": 1.1957,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6684210526315789,
|
| 1791 |
+
"grad_norm": 1.344430318965404,
|
| 1792 |
+
"learning_rate": 5.983045753470308e-06,
|
| 1793 |
+
"loss": 1.2162,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6710526315789473,
|
| 1798 |
+
"grad_norm": 1.5136161435083522,
|
| 1799 |
+
"learning_rate": 5.8990939940156e-06,
|
| 1800 |
+
"loss": 1.1642,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6736842105263158,
|
| 1805 |
+
"grad_norm": 1.362924130944701,
|
| 1806 |
+
"learning_rate": 5.815488272960388e-06,
|
| 1807 |
+
"loss": 1.1774,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6763157894736842,
|
| 1812 |
+
"grad_norm": 1.5146999722224046,
|
| 1813 |
+
"learning_rate": 5.732235645035964e-06,
|
| 1814 |
+
"loss": 1.1654,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6789473684210526,
|
| 1819 |
+
"grad_norm": 1.4713789753634803,
|
| 1820 |
+
"learning_rate": 5.649343135179271e-06,
|
| 1821 |
+
"loss": 1.1692,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6815789473684211,
|
| 1826 |
+
"grad_norm": 1.488147057663755,
|
| 1827 |
+
"learning_rate": 5.566817737940142e-06,
|
| 1828 |
+
"loss": 1.1849,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6842105263157895,
|
| 1833 |
+
"grad_norm": 1.4591392264687968,
|
| 1834 |
+
"learning_rate": 5.484666416891109e-06,
|
| 1835 |
+
"loss": 1.1672,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6868421052631579,
|
| 1840 |
+
"grad_norm": 1.5186815097126352,
|
| 1841 |
+
"learning_rate": 5.4028961040397765e-06,
|
| 1842 |
+
"loss": 1.1856,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6894736842105263,
|
| 1847 |
+
"grad_norm": 1.3841513849936795,
|
| 1848 |
+
"learning_rate": 5.321513699243924e-06,
|
| 1849 |
+
"loss": 1.1278,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6921052631578948,
|
| 1854 |
+
"grad_norm": 1.427452900119818,
|
| 1855 |
+
"learning_rate": 5.240526069629265e-06,
|
| 1856 |
+
"loss": 1.1377,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6947368421052632,
|
| 1861 |
+
"grad_norm": 1.4548570509065128,
|
| 1862 |
+
"learning_rate": 5.159940049010015e-06,
|
| 1863 |
+
"loss": 1.1701,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6973684210526315,
|
| 1868 |
+
"grad_norm": 1.4097905499526544,
|
| 1869 |
+
"learning_rate": 5.079762437312219e-06,
|
| 1870 |
+
"loss": 1.188,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7,
|
| 1875 |
+
"grad_norm": 1.4153566806677924,
|
| 1876 |
+
"learning_rate": 5.000000000000003e-06,
|
| 1877 |
+
"loss": 1.1467,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7026315789473684,
|
| 1882 |
+
"grad_norm": 1.4520190387088883,
|
| 1883 |
+
"learning_rate": 4.9206594675046595e-06,
|
| 1884 |
+
"loss": 1.1669,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7052631578947368,
|
| 1889 |
+
"grad_norm": 1.4607003414092599,
|
| 1890 |
+
"learning_rate": 4.8417475346567635e-06,
|
| 1891 |
+
"loss": 1.2034,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7078947368421052,
|
| 1896 |
+
"grad_norm": 1.5189089646509135,
|
| 1897 |
+
"learning_rate": 4.763270860121222e-06,
|
| 1898 |
+
"loss": 1.1853,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7105263157894737,
|
| 1903 |
+
"grad_norm": 1.5001504965368353,
|
| 1904 |
+
"learning_rate": 4.685236065835443e-06,
|
| 1905 |
+
"loss": 1.1626,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7131578947368421,
|
| 1910 |
+
"grad_norm": 1.4272443913750283,
|
| 1911 |
+
"learning_rate": 4.607649736450539e-06,
|
| 1912 |
+
"loss": 1.1903,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7157894736842105,
|
| 1917 |
+
"grad_norm": 1.3923865065357075,
|
| 1918 |
+
"learning_rate": 4.530518418775734e-06,
|
| 1919 |
+
"loss": 1.1625,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.718421052631579,
|
| 1924 |
+
"grad_norm": 1.4677018709010743,
|
| 1925 |
+
"learning_rate": 4.453848621225913e-06,
|
| 1926 |
+
"loss": 1.122,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7210526315789474,
|
| 1931 |
+
"grad_norm": 1.5389664942910415,
|
| 1932 |
+
"learning_rate": 4.3776468132724605e-06,
|
| 1933 |
+
"loss": 1.1395,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7236842105263158,
|
| 1938 |
+
"grad_norm": 1.4779444930355223,
|
| 1939 |
+
"learning_rate": 4.301919424897339e-06,
|
| 1940 |
+
"loss": 1.1765,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7263157894736842,
|
| 1945 |
+
"grad_norm": 1.4120873508728784,
|
| 1946 |
+
"learning_rate": 4.226672846050538e-06,
|
| 1947 |
+
"loss": 1.1649,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7289473684210527,
|
| 1952 |
+
"grad_norm": 1.4481405920413495,
|
| 1953 |
+
"learning_rate": 4.151913426110864e-06,
|
| 1954 |
+
"loss": 1.1813,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7315789473684211,
|
| 1959 |
+
"grad_norm": 1.4110583437714723,
|
| 1960 |
+
"learning_rate": 4.077647473350201e-06,
|
| 1961 |
+
"loss": 1.1792,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7342105263157894,
|
| 1966 |
+
"grad_norm": 1.3998861268309197,
|
| 1967 |
+
"learning_rate": 4.003881254401183e-06,
|
| 1968 |
+
"loss": 1.1795,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7368421052631579,
|
| 1973 |
+
"grad_norm": 1.5389428254089865,
|
| 1974 |
+
"learning_rate": 3.930620993728434e-06,
|
| 1975 |
+
"loss": 1.1761,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7394736842105263,
|
| 1980 |
+
"grad_norm": 1.351873475574077,
|
| 1981 |
+
"learning_rate": 3.857872873103322e-06,
|
| 1982 |
+
"loss": 1.0988,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7421052631578947,
|
| 1987 |
+
"grad_norm": 2.0822849643846864,
|
| 1988 |
+
"learning_rate": 3.7856430310823546e-06,
|
| 1989 |
+
"loss": 1.1535,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7447368421052631,
|
| 1994 |
+
"grad_norm": 1.4774710448410524,
|
| 1995 |
+
"learning_rate": 3.7139375624891795e-06,
|
| 1996 |
+
"loss": 1.1669,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7473684210526316,
|
| 2001 |
+
"grad_norm": 1.3660260513616105,
|
| 2002 |
+
"learning_rate": 3.6427625179003223e-06,
|
| 2003 |
+
"loss": 1.2248,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.75,
|
| 2008 |
+
"grad_norm": 1.500711580269001,
|
| 2009 |
+
"learning_rate": 3.5721239031346067e-06,
|
| 2010 |
+
"loss": 1.177,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7526315789473684,
|
| 2015 |
+
"grad_norm": 1.4652678196971247,
|
| 2016 |
+
"learning_rate": 3.5020276787464058e-06,
|
| 2017 |
+
"loss": 1.1948,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7552631578947369,
|
| 2022 |
+
"grad_norm": 1.4313428064366354,
|
| 2023 |
+
"learning_rate": 3.4324797595226567e-06,
|
| 2024 |
+
"loss": 1.1299,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7578947368421053,
|
| 2029 |
+
"grad_norm": 1.442031209959671,
|
| 2030 |
+
"learning_rate": 3.3634860139837877e-06,
|
| 2031 |
+
"loss": 1.191,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7605263157894737,
|
| 2036 |
+
"grad_norm": 1.3495347644283935,
|
| 2037 |
+
"learning_rate": 3.2950522638885106e-06,
|
| 2038 |
+
"loss": 1.1744,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7631578947368421,
|
| 2043 |
+
"grad_norm": 1.4953661772700806,
|
| 2044 |
+
"learning_rate": 3.2271842837425917e-06,
|
| 2045 |
+
"loss": 1.1923,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7657894736842106,
|
| 2050 |
+
"grad_norm": 1.379584892765396,
|
| 2051 |
+
"learning_rate": 3.1598878003115694e-06,
|
| 2052 |
+
"loss": 1.1936,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7684210526315789,
|
| 2057 |
+
"grad_norm": 1.4357267033014562,
|
| 2058 |
+
"learning_rate": 3.0931684921375572e-06,
|
| 2059 |
+
"loss": 1.1581,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7710526315789473,
|
| 2064 |
+
"grad_norm": 1.4032602137615429,
|
| 2065 |
+
"learning_rate": 3.0270319890600465e-06,
|
| 2066 |
+
"loss": 1.195,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7736842105263158,
|
| 2071 |
+
"grad_norm": 1.3767404690973988,
|
| 2072 |
+
"learning_rate": 2.9614838717408866e-06,
|
| 2073 |
+
"loss": 1.1404,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7763157894736842,
|
| 2078 |
+
"grad_norm": 1.345562765951157,
|
| 2079 |
+
"learning_rate": 2.89652967119336e-06,
|
| 2080 |
+
"loss": 1.1435,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7789473684210526,
|
| 2085 |
+
"grad_norm": 1.3646022188703086,
|
| 2086 |
+
"learning_rate": 2.8321748683154893e-06,
|
| 2087 |
+
"loss": 1.1675,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.781578947368421,
|
| 2092 |
+
"grad_norm": 1.3246631802990927,
|
| 2093 |
+
"learning_rate": 2.7684248934275327e-06,
|
| 2094 |
+
"loss": 1.134,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7842105263157895,
|
| 2099 |
+
"grad_norm": 1.5259313637971876,
|
| 2100 |
+
"learning_rate": 2.7052851258137936e-06,
|
| 2101 |
+
"loss": 1.1572,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7868421052631579,
|
| 2106 |
+
"grad_norm": 1.4588725449617848,
|
| 2107 |
+
"learning_rate": 2.642760893268684e-06,
|
| 2108 |
+
"loss": 1.0957,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7894736842105263,
|
| 2113 |
+
"grad_norm": 1.4085665694169516,
|
| 2114 |
+
"learning_rate": 2.580857471647186e-06,
|
| 2115 |
+
"loss": 1.1724,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.7921052631578948,
|
| 2120 |
+
"grad_norm": 1.4237214859236806,
|
| 2121 |
+
"learning_rate": 2.519580084419646e-06,
|
| 2122 |
+
"loss": 1.1569,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.7947368421052632,
|
| 2127 |
+
"grad_norm": 1.4112079099940082,
|
| 2128 |
+
"learning_rate": 2.4589339022310386e-06,
|
| 2129 |
+
"loss": 1.1308,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.7973684210526316,
|
| 2134 |
+
"grad_norm": 1.4559794115227898,
|
| 2135 |
+
"learning_rate": 2.3989240424646355e-06,
|
| 2136 |
+
"loss": 1.1536,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8,
|
| 2141 |
+
"grad_norm": 1.3655004847316077,
|
| 2142 |
+
"learning_rate": 2.339555568810221e-06,
|
| 2143 |
+
"loss": 1.1168,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8026315789473685,
|
| 2148 |
+
"grad_norm": 1.4070429304898757,
|
| 2149 |
+
"learning_rate": 2.2808334908367914e-06,
|
| 2150 |
+
"loss": 1.1085,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.8052631578947368,
|
| 2155 |
+
"grad_norm": 1.4528346751525203,
|
| 2156 |
+
"learning_rate": 2.2227627635698624e-06,
|
| 2157 |
+
"loss": 1.1411,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.8078947368421052,
|
| 2162 |
+
"grad_norm": 1.4619781742001423,
|
| 2163 |
+
"learning_rate": 2.165348287073339e-06,
|
| 2164 |
+
"loss": 1.1945,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8105263157894737,
|
| 2169 |
+
"grad_norm": 1.4018656275095926,
|
| 2170 |
+
"learning_rate": 2.1085949060360654e-06,
|
| 2171 |
+
"loss": 1.1488,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8131578947368421,
|
| 2176 |
+
"grad_norm": 1.4294429101475141,
|
| 2177 |
+
"learning_rate": 2.052507409363004e-06,
|
| 2178 |
+
"loss": 1.1323,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8157894736842105,
|
| 2183 |
+
"grad_norm": 1.400567639105242,
|
| 2184 |
+
"learning_rate": 1.9970905297711606e-06,
|
| 2185 |
+
"loss": 1.1744,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8184210526315789,
|
| 2190 |
+
"grad_norm": 1.408857813703171,
|
| 2191 |
+
"learning_rate": 1.9423489433902186e-06,
|
| 2192 |
+
"loss": 1.1603,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8210526315789474,
|
| 2197 |
+
"grad_norm": 1.40296491225255,
|
| 2198 |
+
"learning_rate": 1.8882872693679787e-06,
|
| 2199 |
+
"loss": 1.1483,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8236842105263158,
|
| 2204 |
+
"grad_norm": 1.3612503327875418,
|
| 2205 |
+
"learning_rate": 1.8349100694805711e-06,
|
| 2206 |
+
"loss": 1.1416,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8263157894736842,
|
| 2211 |
+
"grad_norm": 1.3364273351895348,
|
| 2212 |
+
"learning_rate": 1.7822218477475496e-06,
|
| 2213 |
+
"loss": 1.128,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8289473684210527,
|
| 2218 |
+
"grad_norm": 1.3850144107615145,
|
| 2219 |
+
"learning_rate": 1.7302270500518181e-06,
|
| 2220 |
+
"loss": 1.1767,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8315789473684211,
|
| 2225 |
+
"grad_norm": 1.4137419991959699,
|
| 2226 |
+
"learning_rate": 1.6789300637645e-06,
|
| 2227 |
+
"loss": 1.1233,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8342105263157895,
|
| 2232 |
+
"grad_norm": 1.4262918205450703,
|
| 2233 |
+
"learning_rate": 1.6283352173747148e-06,
|
| 2234 |
+
"loss": 1.1457,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8368421052631579,
|
| 2239 |
+
"grad_norm": 1.3477003716799063,
|
| 2240 |
+
"learning_rate": 1.578446780124344e-06,
|
| 2241 |
+
"loss": 1.1403,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8394736842105263,
|
| 2246 |
+
"grad_norm": 1.622769299529781,
|
| 2247 |
+
"learning_rate": 1.5292689616477808e-06,
|
| 2248 |
+
"loss": 1.1348,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8421052631578947,
|
| 2253 |
+
"grad_norm": 1.6481690570937264,
|
| 2254 |
+
"learning_rate": 1.4808059116167306e-06,
|
| 2255 |
+
"loss": 1.1258,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.8447368421052631,
|
| 2260 |
+
"grad_norm": 1.5428571262895439,
|
| 2261 |
+
"learning_rate": 1.4330617193900365e-06,
|
| 2262 |
+
"loss": 1.1397,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8473684210526315,
|
| 2267 |
+
"grad_norm": 1.3637682771315494,
|
| 2268 |
+
"learning_rate": 1.3860404136686411e-06,
|
| 2269 |
+
"loss": 1.1371,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.85,
|
| 2274 |
+
"grad_norm": 1.3583297786593427,
|
| 2275 |
+
"learning_rate": 1.339745962155613e-06,
|
| 2276 |
+
"loss": 1.1584,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8526315789473684,
|
| 2281 |
+
"grad_norm": 1.4145815499277343,
|
| 2282 |
+
"learning_rate": 1.294182271221377e-06,
|
| 2283 |
+
"loss": 1.1708,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8552631578947368,
|
| 2288 |
+
"grad_norm": 1.3721640621198408,
|
| 2289 |
+
"learning_rate": 1.2493531855740626e-06,
|
| 2290 |
+
"loss": 1.1529,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8578947368421053,
|
| 2295 |
+
"grad_norm": 1.352658945488651,
|
| 2296 |
+
"learning_rate": 1.2052624879351105e-06,
|
| 2297 |
+
"loss": 1.1646,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8605263157894737,
|
| 2302 |
+
"grad_norm": 1.3623686161121,
|
| 2303 |
+
"learning_rate": 1.1619138987200562e-06,
|
| 2304 |
+
"loss": 1.1597,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8631578947368421,
|
| 2309 |
+
"grad_norm": 1.4297195595936056,
|
| 2310 |
+
"learning_rate": 1.1193110757246251e-06,
|
| 2311 |
+
"loss": 1.1365,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8657894736842106,
|
| 2316 |
+
"grad_norm": 1.3650877614146917,
|
| 2317 |
+
"learning_rate": 1.0774576138160596e-06,
|
| 2318 |
+
"loss": 1.1343,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.868421052631579,
|
| 2323 |
+
"grad_norm": 1.4235475016088133,
|
| 2324 |
+
"learning_rate": 1.0363570446297999e-06,
|
| 2325 |
+
"loss": 1.1278,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8710526315789474,
|
| 2330 |
+
"grad_norm": 1.4372338077155216,
|
| 2331 |
+
"learning_rate": 9.960128362714637e-07,
|
| 2332 |
+
"loss": 1.1742,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8736842105263158,
|
| 2337 |
+
"grad_norm": 1.5566455926553144,
|
| 2338 |
+
"learning_rate": 9.564283930242258e-07,
|
| 2339 |
+
"loss": 1.1438,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8763157894736842,
|
| 2344 |
+
"grad_norm": 1.4182208939245209,
|
| 2345 |
+
"learning_rate": 9.176070550615379e-07,
|
| 2346 |
+
"loss": 1.0908,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8789473684210526,
|
| 2351 |
+
"grad_norm": 1.3665860472536333,
|
| 2352 |
+
"learning_rate": 8.79552098165296e-07,
|
| 2353 |
+
"loss": 1.1923,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.881578947368421,
|
| 2358 |
+
"grad_norm": 1.4802127975920716,
|
| 2359 |
+
"learning_rate": 8.42266733449425e-07,
|
| 2360 |
+
"loss": 1.1677,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8842105263157894,
|
| 2365 |
+
"grad_norm": 1.3840701915014118,
|
| 2366 |
+
"learning_rate": 8.057541070889229e-07,
|
| 2367 |
+
"loss": 1.1798,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.8868421052631579,
|
| 2372 |
+
"grad_norm": 1.4442257549760622,
|
| 2373 |
+
"learning_rate": 7.700173000543742e-07,
|
| 2374 |
+
"loss": 1.1337,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.8894736842105263,
|
| 2379 |
+
"grad_norm": 1.4991615103817875,
|
| 2380 |
+
"learning_rate": 7.350593278519824e-07,
|
| 2381 |
+
"loss": 1.1288,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.8921052631578947,
|
| 2386 |
+
"grad_norm": 1.3415829958596235,
|
| 2387 |
+
"learning_rate": 7.00883140269123e-07,
|
| 2388 |
+
"loss": 1.1375,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8947368421052632,
|
| 2393 |
+
"grad_norm": 1.4310174671778038,
|
| 2394 |
+
"learning_rate": 6.67491621125429e-07,
|
| 2395 |
+
"loss": 1.1315,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.8973684210526316,
|
| 2400 |
+
"grad_norm": 1.4693458756161413,
|
| 2401 |
+
"learning_rate": 6.348875880294536e-07,
|
| 2402 |
+
"loss": 1.1743,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9,
|
| 2407 |
+
"grad_norm": 1.514756438197858,
|
| 2408 |
+
"learning_rate": 6.030737921409169e-07,
|
| 2409 |
+
"loss": 1.1522,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9026315789473685,
|
| 2414 |
+
"grad_norm": 1.3943946153157085,
|
| 2415 |
+
"learning_rate": 5.720529179385659e-07,
|
| 2416 |
+
"loss": 1.1513,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9052631578947369,
|
| 2421 |
+
"grad_norm": 1.3760568727093738,
|
| 2422 |
+
"learning_rate": 5.418275829936537e-07,
|
| 2423 |
+
"loss": 1.133,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.9078947368421053,
|
| 2428 |
+
"grad_norm": 1.4135104685753546,
|
| 2429 |
+
"learning_rate": 5.124003377490582e-07,
|
| 2430 |
+
"loss": 1.1218,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.9105263157894737,
|
| 2435 |
+
"grad_norm": 1.4715025682557128,
|
| 2436 |
+
"learning_rate": 4.837736653040825e-07,
|
| 2437 |
+
"loss": 1.1973,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.9131578947368421,
|
| 2442 |
+
"grad_norm": 1.365927408406846,
|
| 2443 |
+
"learning_rate": 4.5594998120492505e-07,
|
| 2444 |
+
"loss": 1.1536,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9157894736842105,
|
| 2449 |
+
"grad_norm": 1.395477739011334,
|
| 2450 |
+
"learning_rate": 4.2893163324085886e-07,
|
| 2451 |
+
"loss": 1.1615,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9184210526315789,
|
| 2456 |
+
"grad_norm": 1.3927780195869306,
|
| 2457 |
+
"learning_rate": 4.0272090124611086e-07,
|
| 2458 |
+
"loss": 1.1419,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9210526315789473,
|
| 2463 |
+
"grad_norm": 1.3763363664716992,
|
| 2464 |
+
"learning_rate": 3.773199969074959e-07,
|
| 2465 |
+
"loss": 1.1469,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9236842105263158,
|
| 2470 |
+
"grad_norm": 2.094668256525704,
|
| 2471 |
+
"learning_rate": 3.5273106357779585e-07,
|
| 2472 |
+
"loss": 1.147,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9263157894736842,
|
| 2477 |
+
"grad_norm": 1.4485811313077628,
|
| 2478 |
+
"learning_rate": 3.2895617609489337e-07,
|
| 2479 |
+
"loss": 1.1192,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9289473684210526,
|
| 2484 |
+
"grad_norm": 1.3342469890527902,
|
| 2485 |
+
"learning_rate": 3.059973406066963e-07,
|
| 2486 |
+
"loss": 1.1634,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9315789473684211,
|
| 2491 |
+
"grad_norm": 1.4601464593441522,
|
| 2492 |
+
"learning_rate": 2.838564944018618e-07,
|
| 2493 |
+
"loss": 1.1391,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9342105263157895,
|
| 2498 |
+
"grad_norm": 1.3496811554754882,
|
| 2499 |
+
"learning_rate": 2.62535505746323e-07,
|
| 2500 |
+
"loss": 1.1296,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9368421052631579,
|
| 2505 |
+
"grad_norm": 1.325046815616541,
|
| 2506 |
+
"learning_rate": 2.420361737256438e-07,
|
| 2507 |
+
"loss": 1.1632,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9394736842105263,
|
| 2512 |
+
"grad_norm": 1.3476423015496304,
|
| 2513 |
+
"learning_rate": 2.22360228093208e-07,
|
| 2514 |
+
"loss": 1.147,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9421052631578948,
|
| 2519 |
+
"grad_norm": 1.7118305115455923,
|
| 2520 |
+
"learning_rate": 2.035093291242607e-07,
|
| 2521 |
+
"loss": 1.1268,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9447368421052632,
|
| 2526 |
+
"grad_norm": 1.3942000674318626,
|
| 2527 |
+
"learning_rate": 1.854850674758213e-07,
|
| 2528 |
+
"loss": 1.1581,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9473684210526315,
|
| 2533 |
+
"grad_norm": 1.376220680811591,
|
| 2534 |
+
"learning_rate": 1.6828896405244988e-07,
|
| 2535 |
+
"loss": 1.1295,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.95,
|
| 2540 |
+
"grad_norm": 1.3431256769569868,
|
| 2541 |
+
"learning_rate": 1.519224698779198e-07,
|
| 2542 |
+
"loss": 1.0882,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9526315789473684,
|
| 2547 |
+
"grad_norm": 1.5986582331792596,
|
| 2548 |
+
"learning_rate": 1.3638696597277678e-07,
|
| 2549 |
+
"loss": 1.1812,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9552631578947368,
|
| 2554 |
+
"grad_norm": 1.4132092269970191,
|
| 2555 |
+
"learning_rate": 1.2168376323780652e-07,
|
| 2556 |
+
"loss": 1.1915,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9578947368421052,
|
| 2561 |
+
"grad_norm": 1.430700753393397,
|
| 2562 |
+
"learning_rate": 1.0781410234342093e-07,
|
| 2563 |
+
"loss": 1.1559,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9605263157894737,
|
| 2568 |
+
"grad_norm": 1.3578426630468123,
|
| 2569 |
+
"learning_rate": 9.47791536249676e-08,
|
| 2570 |
+
"loss": 1.1253,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9631578947368421,
|
| 2575 |
+
"grad_norm": 1.34199016293369,
|
| 2576 |
+
"learning_rate": 8.258001698397744e-08,
|
| 2577 |
+
"loss": 1.1275,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9657894736842105,
|
| 2582 |
+
"grad_norm": 1.3736910582430197,
|
| 2583 |
+
"learning_rate": 7.121772179535135e-08,
|
| 2584 |
+
"loss": 1.1552,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.968421052631579,
|
| 2589 |
+
"grad_norm": 1.4235287172921616,
|
| 2590 |
+
"learning_rate": 6.069322682050516e-08,
|
| 2591 |
+
"loss": 1.1396,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9710526315789474,
|
| 2596 |
+
"grad_norm": 1.5091671248608574,
|
| 2597 |
+
"learning_rate": 5.10074201264632e-08,
|
| 2598 |
+
"loss": 1.1243,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9736842105263158,
|
| 2603 |
+
"grad_norm": 1.3411846884431124,
|
| 2604 |
+
"learning_rate": 4.216111901092501e-08,
|
| 2605 |
+
"loss": 1.144,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9763157894736842,
|
| 2610 |
+
"grad_norm": 1.4483679456926384,
|
| 2611 |
+
"learning_rate": 3.4155069933301535e-08,
|
| 2612 |
+
"loss": 1.1223,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.9789473684210527,
|
| 2617 |
+
"grad_norm": 1.395711615853488,
|
| 2618 |
+
"learning_rate": 2.6989948451726643e-08,
|
| 2619 |
+
"loss": 1.1685,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9815789473684211,
|
| 2624 |
+
"grad_norm": 1.3902035335632743,
|
| 2625 |
+
"learning_rate": 2.066635916605386e-08,
|
| 2626 |
+
"loss": 1.0879,
|
| 2627 |
+
"step": 1865
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.9842105263157894,
|
| 2631 |
+
"grad_norm": 1.4063494990348104,
|
| 2632 |
+
"learning_rate": 1.518483566683826e-08,
|
| 2633 |
+
"loss": 1.1322,
|
| 2634 |
+
"step": 1870
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.9868421052631579,
|
| 2638 |
+
"grad_norm": 1.3644257494193968,
|
| 2639 |
+
"learning_rate": 1.0545840490313597e-08,
|
| 2640 |
+
"loss": 1.1111,
|
| 2641 |
+
"step": 1875
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.9894736842105263,
|
| 2645 |
+
"grad_norm": 1.3443180245395077,
|
| 2646 |
+
"learning_rate": 6.749765079363535e-09,
|
| 2647 |
+
"loss": 1.1258,
|
| 2648 |
+
"step": 1880
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.9921052631578947,
|
| 2652 |
+
"grad_norm": 1.4222815730212364,
|
| 2653 |
+
"learning_rate": 3.7969297504858445e-09,
|
| 2654 |
+
"loss": 1.15,
|
| 2655 |
+
"step": 1885
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.9947368421052631,
|
| 2659 |
+
"grad_norm": 1.4727085543827443,
|
| 2660 |
+
"learning_rate": 1.6875836667729073e-09,
|
| 2661 |
+
"loss": 1.1451,
|
| 2662 |
+
"step": 1890
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.9973684210526316,
|
| 2666 |
+
"grad_norm": 1.3787627195454673,
|
| 2667 |
+
"learning_rate": 4.2190481687631736e-10,
|
| 2668 |
+
"loss": 1.1622,
|
| 2669 |
+
"step": 1895
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 1.0,
|
| 2673 |
+
"grad_norm": 1.4191400815453412,
|
| 2674 |
+
"learning_rate": 0.0,
|
| 2675 |
+
"loss": 1.0871,
|
| 2676 |
+
"step": 1900
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 1.0,
|
| 2680 |
+
"eval_loss": 1.1701046228408813,
|
| 2681 |
+
"eval_runtime": 1144.8239,
|
| 2682 |
+
"eval_samples_per_second": 23.514,
|
| 2683 |
+
"eval_steps_per_second": 0.735,
|
| 2684 |
+
"step": 1900
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 1.0,
|
| 2688 |
+
"step": 1900,
|
| 2689 |
+
"total_flos": 99455336448000.0,
|
| 2690 |
+
"train_loss": 1.1960604431754664,
|
| 2691 |
+
"train_runtime": 10060.3773,
|
| 2692 |
+
"train_samples_per_second": 6.043,
|
| 2693 |
+
"train_steps_per_second": 0.189
|
| 2694 |
+
}
|
| 2695 |
+
],
|
| 2696 |
+
"logging_steps": 5,
|
| 2697 |
+
"max_steps": 1900,
|
| 2698 |
+
"num_input_tokens_seen": 0,
|
| 2699 |
+
"num_train_epochs": 1,
|
| 2700 |
+
"save_steps": 500,
|
| 2701 |
+
"stateful_callbacks": {
|
| 2702 |
+
"TrainerControl": {
|
| 2703 |
+
"args": {
|
| 2704 |
+
"should_epoch_stop": false,
|
| 2705 |
+
"should_evaluate": false,
|
| 2706 |
+
"should_log": false,
|
| 2707 |
+
"should_save": true,
|
| 2708 |
+
"should_training_stop": true
|
| 2709 |
+
},
|
| 2710 |
+
"attributes": {}
|
| 2711 |
+
}
|
| 2712 |
+
},
|
| 2713 |
+
"total_flos": 99455336448000.0,
|
| 2714 |
+
"train_batch_size": 8,
|
| 2715 |
+
"trial_name": null,
|
| 2716 |
+
"trial_params": null
|
| 2717 |
+
}
|