Commit
•
77fc07b
1
Parent(s):
20a11fd
Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- generation_config.json +9 -0
- runs/Nov04_20-39-14_f9834025b1f4/events.out.tfevents.1730753503.f9834025b1f4.3326.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2745 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Llama-3.2-3B
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: llama3.2-3B-cpt-1B-tokens
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for llama3.2-3B-cpt-1B-tokens
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B).
|
| 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="skymizer/llama3.2-3B-cpt-1B-tokens", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/eli3051/huggingface/runs/9pax5l0q)
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.0
|
| 37 |
+
- Transformers: 4.46.1
|
| 38 |
+
- Pytorch: 2.4.0
|
| 39 |
+
- Datasets: 3.1.0
|
| 40 |
+
- Tokenizers: 0.20.2
|
| 41 |
+
|
| 42 |
+
## Citations
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
Cite TRL as:
|
| 47 |
+
|
| 48 |
+
```bibtex
|
| 49 |
+
@misc{vonwerra2022trl,
|
| 50 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 51 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
|
| 52 |
+
year = 2020,
|
| 53 |
+
journal = {GitHub repository},
|
| 54 |
+
publisher = {GitHub},
|
| 55 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 56 |
+
}
|
| 57 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.999739921976593,
|
| 3 |
+
"total_flos": 1058556698689536.0,
|
| 4 |
+
"train_loss": 2.26795463333765,
|
| 5 |
+
"train_runtime": 9902.9795,
|
| 6 |
+
"train_samples": 1250000,
|
| 7 |
+
"train_samples_per_second": 49.694,
|
| 8 |
+
"train_steps_per_second": 0.194
|
| 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": "4.46.1"
|
| 9 |
+
}
|
runs/Nov04_20-39-14_f9834025b1f4/events.out.tfevents.1730753503.f9834025b1f4.3326.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29b7f9543e268960febd8687f86cb3da4208e6005fc691181b685baf674e4c6a
|
| 3 |
+
size 87710
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.999739921976593,
|
| 3 |
+
"total_flos": 1058556698689536.0,
|
| 4 |
+
"train_loss": 2.26795463333765,
|
| 5 |
+
"train_runtime": 9902.9795,
|
| 6 |
+
"train_samples": 1250000,
|
| 7 |
+
"train_samples_per_second": 49.694,
|
| 8 |
+
"train_steps_per_second": 0.194
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2745 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.999739921976593,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1922,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005201560468140442,
|
| 13 |
+
"grad_norm": 0.8345186041103929,
|
| 14 |
+
"learning_rate": 1.0362694300518136e-07,
|
| 15 |
+
"loss": 2.3004,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.002600780234070221,
|
| 20 |
+
"grad_norm": 0.7480244688392631,
|
| 21 |
+
"learning_rate": 5.181347150259067e-07,
|
| 22 |
+
"loss": 2.3277,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005201560468140442,
|
| 27 |
+
"grad_norm": 0.785533525256438,
|
| 28 |
+
"learning_rate": 1.0362694300518134e-06,
|
| 29 |
+
"loss": 2.2876,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.007802340702210663,
|
| 34 |
+
"grad_norm": 1.9886611293748466,
|
| 35 |
+
"learning_rate": 1.5544041450777204e-06,
|
| 36 |
+
"loss": 2.2908,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010403120936280884,
|
| 41 |
+
"grad_norm": 0.5867201261604411,
|
| 42 |
+
"learning_rate": 2.072538860103627e-06,
|
| 43 |
+
"loss": 2.3241,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013003901170351105,
|
| 48 |
+
"grad_norm": 0.4918894228225657,
|
| 49 |
+
"learning_rate": 2.5906735751295338e-06,
|
| 50 |
+
"loss": 2.2845,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.015604681404421327,
|
| 55 |
+
"grad_norm": 0.5030371996482103,
|
| 56 |
+
"learning_rate": 3.1088082901554407e-06,
|
| 57 |
+
"loss": 2.2913,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.018205461638491547,
|
| 62 |
+
"grad_norm": 0.4576003124686891,
|
| 63 |
+
"learning_rate": 3.6269430051813476e-06,
|
| 64 |
+
"loss": 2.2546,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.02080624187256177,
|
| 69 |
+
"grad_norm": 0.4539314680873136,
|
| 70 |
+
"learning_rate": 4.145077720207254e-06,
|
| 71 |
+
"loss": 2.2961,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02340702210663199,
|
| 76 |
+
"grad_norm": 0.4401657938174189,
|
| 77 |
+
"learning_rate": 4.663212435233161e-06,
|
| 78 |
+
"loss": 2.2901,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.02600780234070221,
|
| 83 |
+
"grad_norm": 0.43168803583003423,
|
| 84 |
+
"learning_rate": 5.1813471502590676e-06,
|
| 85 |
+
"loss": 2.2724,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.02860858257477243,
|
| 90 |
+
"grad_norm": 0.464515894991219,
|
| 91 |
+
"learning_rate": 5.699481865284975e-06,
|
| 92 |
+
"loss": 2.2706,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.031209362808842653,
|
| 97 |
+
"grad_norm": 0.41118157204418754,
|
| 98 |
+
"learning_rate": 6.217616580310881e-06,
|
| 99 |
+
"loss": 2.2826,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.033810143042912875,
|
| 104 |
+
"grad_norm": 0.41838223457309465,
|
| 105 |
+
"learning_rate": 6.735751295336788e-06,
|
| 106 |
+
"loss": 2.2842,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.036410923276983094,
|
| 111 |
+
"grad_norm": 0.4116710918403529,
|
| 112 |
+
"learning_rate": 7.253886010362695e-06,
|
| 113 |
+
"loss": 2.2675,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.03901170351105332,
|
| 118 |
+
"grad_norm": 0.4294733222487177,
|
| 119 |
+
"learning_rate": 7.772020725388602e-06,
|
| 120 |
+
"loss": 2.2703,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04161248374512354,
|
| 125 |
+
"grad_norm": 0.4118250916081217,
|
| 126 |
+
"learning_rate": 8.290155440414507e-06,
|
| 127 |
+
"loss": 2.2764,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.044213263979193757,
|
| 132 |
+
"grad_norm": 0.41023756575889064,
|
| 133 |
+
"learning_rate": 8.808290155440415e-06,
|
| 134 |
+
"loss": 2.2902,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.04681404421326398,
|
| 139 |
+
"grad_norm": 0.43880292962496026,
|
| 140 |
+
"learning_rate": 9.326424870466322e-06,
|
| 141 |
+
"loss": 2.2912,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0494148244473342,
|
| 146 |
+
"grad_norm": 0.45031096977273377,
|
| 147 |
+
"learning_rate": 9.844559585492228e-06,
|
| 148 |
+
"loss": 2.2981,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05201560468140442,
|
| 153 |
+
"grad_norm": 0.4248849523300326,
|
| 154 |
+
"learning_rate": 1.0362694300518135e-05,
|
| 155 |
+
"loss": 2.2975,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.054616384915474644,
|
| 160 |
+
"grad_norm": 0.41266838355458524,
|
| 161 |
+
"learning_rate": 1.0880829015544042e-05,
|
| 162 |
+
"loss": 2.2736,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05721716514954486,
|
| 167 |
+
"grad_norm": 0.4094053779887782,
|
| 168 |
+
"learning_rate": 1.139896373056995e-05,
|
| 169 |
+
"loss": 2.255,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.05981794538361508,
|
| 174 |
+
"grad_norm": 0.4136951723970802,
|
| 175 |
+
"learning_rate": 1.1917098445595855e-05,
|
| 176 |
+
"loss": 2.2758,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06241872561768531,
|
| 181 |
+
"grad_norm": 0.43700184963849786,
|
| 182 |
+
"learning_rate": 1.2435233160621763e-05,
|
| 183 |
+
"loss": 2.3071,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06501950585175553,
|
| 188 |
+
"grad_norm": 0.411543204357628,
|
| 189 |
+
"learning_rate": 1.2953367875647668e-05,
|
| 190 |
+
"loss": 2.2826,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06762028608582575,
|
| 195 |
+
"grad_norm": 0.429531396061487,
|
| 196 |
+
"learning_rate": 1.3471502590673576e-05,
|
| 197 |
+
"loss": 2.2612,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07022106631989597,
|
| 202 |
+
"grad_norm": 0.4137946537504368,
|
| 203 |
+
"learning_rate": 1.3989637305699483e-05,
|
| 204 |
+
"loss": 2.2724,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07282184655396619,
|
| 209 |
+
"grad_norm": 0.4108339707707568,
|
| 210 |
+
"learning_rate": 1.450777202072539e-05,
|
| 211 |
+
"loss": 2.2743,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.0754226267880364,
|
| 216 |
+
"grad_norm": 0.4212937350493228,
|
| 217 |
+
"learning_rate": 1.5025906735751296e-05,
|
| 218 |
+
"loss": 2.2688,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07802340702210664,
|
| 223 |
+
"grad_norm": 0.42079895435460035,
|
| 224 |
+
"learning_rate": 1.5544041450777204e-05,
|
| 225 |
+
"loss": 2.2881,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08062418725617686,
|
| 230 |
+
"grad_norm": 0.4355924970314073,
|
| 231 |
+
"learning_rate": 1.606217616580311e-05,
|
| 232 |
+
"loss": 2.2937,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08322496749024708,
|
| 237 |
+
"grad_norm": 0.41567781768472056,
|
| 238 |
+
"learning_rate": 1.6580310880829015e-05,
|
| 239 |
+
"loss": 2.2827,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.0858257477243173,
|
| 244 |
+
"grad_norm": 0.4212845025959482,
|
| 245 |
+
"learning_rate": 1.7098445595854924e-05,
|
| 246 |
+
"loss": 2.2754,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.08842652795838751,
|
| 251 |
+
"grad_norm": 0.428136588267208,
|
| 252 |
+
"learning_rate": 1.761658031088083e-05,
|
| 253 |
+
"loss": 2.2915,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09102730819245773,
|
| 258 |
+
"grad_norm": 0.4336880983720985,
|
| 259 |
+
"learning_rate": 1.813471502590674e-05,
|
| 260 |
+
"loss": 2.2406,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09362808842652796,
|
| 265 |
+
"grad_norm": 0.4279056529435773,
|
| 266 |
+
"learning_rate": 1.8652849740932644e-05,
|
| 267 |
+
"loss": 2.2566,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09622886866059818,
|
| 272 |
+
"grad_norm": 0.43624158893620124,
|
| 273 |
+
"learning_rate": 1.917098445595855e-05,
|
| 274 |
+
"loss": 2.2678,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.0988296488946684,
|
| 279 |
+
"grad_norm": 0.4386962113064878,
|
| 280 |
+
"learning_rate": 1.9689119170984456e-05,
|
| 281 |
+
"loss": 2.3003,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10143042912873862,
|
| 286 |
+
"grad_norm": 0.4399748777484219,
|
| 287 |
+
"learning_rate": 1.999993397030723e-05,
|
| 288 |
+
"loss": 2.2561,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10403120936280884,
|
| 293 |
+
"grad_norm": 0.4459322004676816,
|
| 294 |
+
"learning_rate": 1.999919114627769e-05,
|
| 295 |
+
"loss": 2.2792,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.10663198959687907,
|
| 300 |
+
"grad_norm": 0.4787906575891283,
|
| 301 |
+
"learning_rate": 1.9997623022617216e-05,
|
| 302 |
+
"loss": 2.2781,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.10923276983094929,
|
| 307 |
+
"grad_norm": 0.5593634080047786,
|
| 308 |
+
"learning_rate": 1.9995229728753464e-05,
|
| 309 |
+
"loss": 2.2821,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11183355006501951,
|
| 314 |
+
"grad_norm": 0.4556491907298568,
|
| 315 |
+
"learning_rate": 1.9992011462220863e-05,
|
| 316 |
+
"loss": 2.287,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11443433029908973,
|
| 321 |
+
"grad_norm": 0.43460978291949764,
|
| 322 |
+
"learning_rate": 1.9987968488644317e-05,
|
| 323 |
+
"loss": 2.2563,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.11703511053315994,
|
| 328 |
+
"grad_norm": 0.4335991371643843,
|
| 329 |
+
"learning_rate": 1.9983101141717284e-05,
|
| 330 |
+
"loss": 2.257,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.11963589076723016,
|
| 335 |
+
"grad_norm": 0.4224092667055087,
|
| 336 |
+
"learning_rate": 1.9977409823174212e-05,
|
| 337 |
+
"loss": 2.294,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1222366710013004,
|
| 342 |
+
"grad_norm": 0.4485005361456074,
|
| 343 |
+
"learning_rate": 1.9970895002757413e-05,
|
| 344 |
+
"loss": 2.2612,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12483745123537061,
|
| 349 |
+
"grad_norm": 0.4370710376897024,
|
| 350 |
+
"learning_rate": 1.996355721817826e-05,
|
| 351 |
+
"loss": 2.274,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.12743823146944083,
|
| 356 |
+
"grad_norm": 0.44855062619100494,
|
| 357 |
+
"learning_rate": 1.995539707507284e-05,
|
| 358 |
+
"loss": 2.276,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13003901170351106,
|
| 363 |
+
"grad_norm": 1.2608831364594368,
|
| 364 |
+
"learning_rate": 1.9946415246951928e-05,
|
| 365 |
+
"loss": 2.2903,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13263979193758127,
|
| 370 |
+
"grad_norm": 0.4458820122839497,
|
| 371 |
+
"learning_rate": 1.993661247514544e-05,
|
| 372 |
+
"loss": 2.2815,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1352405721716515,
|
| 377 |
+
"grad_norm": 0.42495144352793196,
|
| 378 |
+
"learning_rate": 1.992598956874121e-05,
|
| 379 |
+
"loss": 2.2958,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.1378413524057217,
|
| 384 |
+
"grad_norm": 0.41620906246364725,
|
| 385 |
+
"learning_rate": 1.9914547404518235e-05,
|
| 386 |
+
"loss": 2.2586,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.14044213263979194,
|
| 391 |
+
"grad_norm": 0.43094630659032057,
|
| 392 |
+
"learning_rate": 1.990228692687429e-05,
|
| 393 |
+
"loss": 2.3091,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14304291287386217,
|
| 398 |
+
"grad_norm": 0.42273846563736506,
|
| 399 |
+
"learning_rate": 1.988920914774799e-05,
|
| 400 |
+
"loss": 2.2842,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14564369310793238,
|
| 405 |
+
"grad_norm": 0.4257951665830633,
|
| 406 |
+
"learning_rate": 1.987531514653528e-05,
|
| 407 |
+
"loss": 2.2724,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1482444733420026,
|
| 412 |
+
"grad_norm": 0.44805118487610107,
|
| 413 |
+
"learning_rate": 1.9860606070000318e-05,
|
| 414 |
+
"loss": 2.2645,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1508452535760728,
|
| 419 |
+
"grad_norm": 0.41304563534445005,
|
| 420 |
+
"learning_rate": 1.9845083132180844e-05,
|
| 421 |
+
"loss": 2.2827,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15344603381014305,
|
| 426 |
+
"grad_norm": 0.4317171768002588,
|
| 427 |
+
"learning_rate": 1.982874761428798e-05,
|
| 428 |
+
"loss": 2.2638,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.15604681404421328,
|
| 433 |
+
"grad_norm": 0.43837156488809603,
|
| 434 |
+
"learning_rate": 1.9811600864600463e-05,
|
| 435 |
+
"loss": 2.2958,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.15864759427828348,
|
| 440 |
+
"grad_norm": 0.42186569358732456,
|
| 441 |
+
"learning_rate": 1.979364429835339e-05,
|
| 442 |
+
"loss": 2.2913,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16124837451235371,
|
| 447 |
+
"grad_norm": 0.40765691963435197,
|
| 448 |
+
"learning_rate": 1.9774879397621387e-05,
|
| 449 |
+
"loss": 2.2568,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16384915474642392,
|
| 454 |
+
"grad_norm": 0.42957228218399773,
|
| 455 |
+
"learning_rate": 1.9755307711196296e-05,
|
| 456 |
+
"loss": 2.2671,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.16644993498049415,
|
| 461 |
+
"grad_norm": 0.5983846623121506,
|
| 462 |
+
"learning_rate": 1.973493085445933e-05,
|
| 463 |
+
"loss": 2.2708,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.16905071521456436,
|
| 468 |
+
"grad_norm": 0.46755984948188173,
|
| 469 |
+
"learning_rate": 1.9713750509247766e-05,
|
| 470 |
+
"loss": 2.2565,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.1716514954486346,
|
| 475 |
+
"grad_norm": 0.5074022712445009,
|
| 476 |
+
"learning_rate": 1.969176842371611e-05,
|
| 477 |
+
"loss": 2.2852,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.17425227568270482,
|
| 482 |
+
"grad_norm": 0.4177276291117073,
|
| 483 |
+
"learning_rate": 1.9668986412191822e-05,
|
| 484 |
+
"loss": 2.2674,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.17685305591677503,
|
| 489 |
+
"grad_norm": 0.4451543573816552,
|
| 490 |
+
"learning_rate": 1.9645406355025565e-05,
|
| 491 |
+
"loss": 2.2863,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.17945383615084526,
|
| 496 |
+
"grad_norm": 0.4549495732621677,
|
| 497 |
+
"learning_rate": 1.9621030198436007e-05,
|
| 498 |
+
"loss": 2.2818,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.18205461638491546,
|
| 503 |
+
"grad_norm": 0.43066970227078843,
|
| 504 |
+
"learning_rate": 1.9595859954349187e-05,
|
| 505 |
+
"loss": 2.2715,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.1846553966189857,
|
| 510 |
+
"grad_norm": 0.45210248976895245,
|
| 511 |
+
"learning_rate": 1.956989770023246e-05,
|
| 512 |
+
"loss": 2.2835,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.18725617685305593,
|
| 517 |
+
"grad_norm": 0.43430647379108794,
|
| 518 |
+
"learning_rate": 1.954314557892302e-05,
|
| 519 |
+
"loss": 2.2724,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.18985695708712613,
|
| 524 |
+
"grad_norm": 0.4138706040482473,
|
| 525 |
+
"learning_rate": 1.9515605798451054e-05,
|
| 526 |
+
"loss": 2.2542,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.19245773732119636,
|
| 531 |
+
"grad_norm": 0.41813079660766717,
|
| 532 |
+
"learning_rate": 1.948728063185748e-05,
|
| 533 |
+
"loss": 2.2692,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.19505851755526657,
|
| 538 |
+
"grad_norm": 0.41092129187872495,
|
| 539 |
+
"learning_rate": 1.9458172417006347e-05,
|
| 540 |
+
"loss": 2.3078,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.1976592977893368,
|
| 545 |
+
"grad_norm": 0.4287277843333683,
|
| 546 |
+
"learning_rate": 1.942828355639189e-05,
|
| 547 |
+
"loss": 2.2639,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.20026007802340703,
|
| 552 |
+
"grad_norm": 0.5319302745918535,
|
| 553 |
+
"learning_rate": 1.9397616516940195e-05,
|
| 554 |
+
"loss": 2.2949,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.20286085825747724,
|
| 559 |
+
"grad_norm": 0.4258251973151665,
|
| 560 |
+
"learning_rate": 1.936617382980565e-05,
|
| 561 |
+
"loss": 2.2972,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.20546163849154747,
|
| 566 |
+
"grad_norm": 0.437526049230875,
|
| 567 |
+
"learning_rate": 1.9333958090161955e-05,
|
| 568 |
+
"loss": 2.2405,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.20806241872561768,
|
| 573 |
+
"grad_norm": 0.4363771353732369,
|
| 574 |
+
"learning_rate": 1.9300971956987996e-05,
|
| 575 |
+
"loss": 2.3013,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2106631989596879,
|
| 580 |
+
"grad_norm": 0.43404759577988555,
|
| 581 |
+
"learning_rate": 1.9267218152848342e-05,
|
| 582 |
+
"loss": 2.2843,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.21326397919375814,
|
| 587 |
+
"grad_norm": 0.42684871097670124,
|
| 588 |
+
"learning_rate": 1.9232699463668543e-05,
|
| 589 |
+
"loss": 2.265,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.21586475942782835,
|
| 594 |
+
"grad_norm": 0.40630432938608335,
|
| 595 |
+
"learning_rate": 1.919741873850519e-05,
|
| 596 |
+
"loss": 2.2495,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.21846553966189858,
|
| 601 |
+
"grad_norm": 0.4092728915412203,
|
| 602 |
+
"learning_rate": 1.9161378889310763e-05,
|
| 603 |
+
"loss": 2.2837,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.22106631989596878,
|
| 608 |
+
"grad_norm": 0.4269612586962797,
|
| 609 |
+
"learning_rate": 1.9124582890693285e-05,
|
| 610 |
+
"loss": 2.3053,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.22366710013003901,
|
| 615 |
+
"grad_norm": 0.42780196750395155,
|
| 616 |
+
"learning_rate": 1.908703377967082e-05,
|
| 617 |
+
"loss": 2.2844,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.22626788036410922,
|
| 622 |
+
"grad_norm": 0.43404267438118455,
|
| 623 |
+
"learning_rate": 1.9048734655420792e-05,
|
| 624 |
+
"loss": 2.2557,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.22886866059817945,
|
| 629 |
+
"grad_norm": 0.41555747970189066,
|
| 630 |
+
"learning_rate": 1.900968867902419e-05,
|
| 631 |
+
"loss": 2.2557,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23146944083224968,
|
| 636 |
+
"grad_norm": 0.451526901745645,
|
| 637 |
+
"learning_rate": 1.8969899073204687e-05,
|
| 638 |
+
"loss": 2.2511,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2340702210663199,
|
| 643 |
+
"grad_norm": 0.455010350294312,
|
| 644 |
+
"learning_rate": 1.8929369122062612e-05,
|
| 645 |
+
"loss": 2.2784,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.23667100130039012,
|
| 650 |
+
"grad_norm": 0.42013024506881264,
|
| 651 |
+
"learning_rate": 1.8888102170803914e-05,
|
| 652 |
+
"loss": 2.2873,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.23927178153446033,
|
| 657 |
+
"grad_norm": 0.43230673552727256,
|
| 658 |
+
"learning_rate": 1.8846101625464058e-05,
|
| 659 |
+
"loss": 2.2414,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.24187256176853056,
|
| 664 |
+
"grad_norm": 0.4068180890583648,
|
| 665 |
+
"learning_rate": 1.8803370952626895e-05,
|
| 666 |
+
"loss": 2.2904,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2444733420026008,
|
| 671 |
+
"grad_norm": 0.4179891521528739,
|
| 672 |
+
"learning_rate": 1.875991367913855e-05,
|
| 673 |
+
"loss": 2.3023,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.247074122236671,
|
| 678 |
+
"grad_norm": 0.4095908465134204,
|
| 679 |
+
"learning_rate": 1.8715733391816325e-05,
|
| 680 |
+
"loss": 2.2553,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.24967490247074123,
|
| 685 |
+
"grad_norm": 0.4288772358174505,
|
| 686 |
+
"learning_rate": 1.867083373715264e-05,
|
| 687 |
+
"loss": 2.2897,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.25227568270481143,
|
| 692 |
+
"grad_norm": 0.418665612536374,
|
| 693 |
+
"learning_rate": 1.8625218421014093e-05,
|
| 694 |
+
"loss": 2.3026,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.25487646293888166,
|
| 699 |
+
"grad_norm": 0.4054990412366829,
|
| 700 |
+
"learning_rate": 1.857889120833557e-05,
|
| 701 |
+
"loss": 2.2452,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2574772431729519,
|
| 706 |
+
"grad_norm": 0.4244802845116968,
|
| 707 |
+
"learning_rate": 1.853185592280951e-05,
|
| 708 |
+
"loss": 2.2645,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.26007802340702213,
|
| 713 |
+
"grad_norm": 0.41294094690785654,
|
| 714 |
+
"learning_rate": 1.8484116446570286e-05,
|
| 715 |
+
"loss": 2.2745,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.2626788036410923,
|
| 720 |
+
"grad_norm": 0.4416502443149561,
|
| 721 |
+
"learning_rate": 1.8435676719873828e-05,
|
| 722 |
+
"loss": 2.3012,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.26527958387516254,
|
| 727 |
+
"grad_norm": 0.42276919127148177,
|
| 728 |
+
"learning_rate": 1.8386540740772376e-05,
|
| 729 |
+
"loss": 2.3062,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.26788036410923277,
|
| 734 |
+
"grad_norm": 0.42890188987731626,
|
| 735 |
+
"learning_rate": 1.8336712564784506e-05,
|
| 736 |
+
"loss": 2.2855,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.270481144343303,
|
| 741 |
+
"grad_norm": 0.43944006643762495,
|
| 742 |
+
"learning_rate": 1.828619630456039e-05,
|
| 743 |
+
"loss": 2.2945,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.27308192457737324,
|
| 748 |
+
"grad_norm": 0.39897414033069534,
|
| 749 |
+
"learning_rate": 1.823499612954237e-05,
|
| 750 |
+
"loss": 2.2724,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2756827048114434,
|
| 755 |
+
"grad_norm": 0.42692898380084127,
|
| 756 |
+
"learning_rate": 1.8183116265620814e-05,
|
| 757 |
+
"loss": 2.2751,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.27828348504551365,
|
| 762 |
+
"grad_norm": 0.5774651759629665,
|
| 763 |
+
"learning_rate": 1.8130560994785325e-05,
|
| 764 |
+
"loss": 2.2614,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.2808842652795839,
|
| 769 |
+
"grad_norm": 0.4368497226958751,
|
| 770 |
+
"learning_rate": 1.807733465477133e-05,
|
| 771 |
+
"loss": 2.2752,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2834850455136541,
|
| 776 |
+
"grad_norm": 0.3949944652221127,
|
| 777 |
+
"learning_rate": 1.802344163870204e-05,
|
| 778 |
+
"loss": 2.2747,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.28608582574772434,
|
| 783 |
+
"grad_norm": 0.4231648910835165,
|
| 784 |
+
"learning_rate": 1.7968886394725876e-05,
|
| 785 |
+
"loss": 2.2587,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.2886866059817945,
|
| 790 |
+
"grad_norm": 0.41019516846530374,
|
| 791 |
+
"learning_rate": 1.791367342564932e-05,
|
| 792 |
+
"loss": 2.2687,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.29128738621586475,
|
| 797 |
+
"grad_norm": 0.4183646932964066,
|
| 798 |
+
"learning_rate": 1.785780728856527e-05,
|
| 799 |
+
"loss": 2.2942,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.293888166449935,
|
| 804 |
+
"grad_norm": 0.41427010452675417,
|
| 805 |
+
"learning_rate": 1.7801292594476925e-05,
|
| 806 |
+
"loss": 2.2807,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.2964889466840052,
|
| 811 |
+
"grad_norm": 0.40822217591655924,
|
| 812 |
+
"learning_rate": 1.7744134007917195e-05,
|
| 813 |
+
"loss": 2.2787,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.29908972691807545,
|
| 818 |
+
"grad_norm": 0.43458061917820046,
|
| 819 |
+
"learning_rate": 1.768633624656371e-05,
|
| 820 |
+
"loss": 2.2783,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3016905071521456,
|
| 825 |
+
"grad_norm": 0.41243095488481374,
|
| 826 |
+
"learning_rate": 1.7627904080849436e-05,
|
| 827 |
+
"loss": 2.2527,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.30429128738621586,
|
| 832 |
+
"grad_norm": 0.4282495979925253,
|
| 833 |
+
"learning_rate": 1.7568842333568952e-05,
|
| 834 |
+
"loss": 2.2784,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3068920676202861,
|
| 839 |
+
"grad_norm": 0.45009971762477025,
|
| 840 |
+
"learning_rate": 1.7509155879480376e-05,
|
| 841 |
+
"loss": 2.2261,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3094928478543563,
|
| 846 |
+
"grad_norm": 0.4099231262196378,
|
| 847 |
+
"learning_rate": 1.7448849644903024e-05,
|
| 848 |
+
"loss": 2.2744,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.31209362808842656,
|
| 853 |
+
"grad_norm": 0.39853848132220815,
|
| 854 |
+
"learning_rate": 1.738792860731082e-05,
|
| 855 |
+
"loss": 2.3049,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.31469440832249673,
|
| 860 |
+
"grad_norm": 0.4189298954062125,
|
| 861 |
+
"learning_rate": 1.7326397794921453e-05,
|
| 862 |
+
"loss": 2.2771,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.31729518855656696,
|
| 867 |
+
"grad_norm": 0.4062958057669602,
|
| 868 |
+
"learning_rate": 1.7264262286281388e-05,
|
| 869 |
+
"loss": 2.2854,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3198959687906372,
|
| 874 |
+
"grad_norm": 0.4174075877503918,
|
| 875 |
+
"learning_rate": 1.7201527209846665e-05,
|
| 876 |
+
"loss": 2.3018,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.32249674902470743,
|
| 881 |
+
"grad_norm": 0.40725167976981597,
|
| 882 |
+
"learning_rate": 1.7138197743559656e-05,
|
| 883 |
+
"loss": 2.2715,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.3250975292587776,
|
| 888 |
+
"grad_norm": 0.43501733923677793,
|
| 889 |
+
"learning_rate": 1.7074279114421655e-05,
|
| 890 |
+
"loss": 2.2777,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.32769830949284784,
|
| 895 |
+
"grad_norm": 0.4267668851051862,
|
| 896 |
+
"learning_rate": 1.7009776598061496e-05,
|
| 897 |
+
"loss": 2.2632,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.33029908972691807,
|
| 902 |
+
"grad_norm": 0.41325763071560917,
|
| 903 |
+
"learning_rate": 1.6944695518300087e-05,
|
| 904 |
+
"loss": 2.2961,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3328998699609883,
|
| 909 |
+
"grad_norm": 0.4097349085762526,
|
| 910 |
+
"learning_rate": 1.6879041246711025e-05,
|
| 911 |
+
"loss": 2.2772,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.33550065019505854,
|
| 916 |
+
"grad_norm": 0.4118740879092736,
|
| 917 |
+
"learning_rate": 1.6812819202177225e-05,
|
| 918 |
+
"loss": 2.2927,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3381014304291287,
|
| 923 |
+
"grad_norm": 0.40576237897775325,
|
| 924 |
+
"learning_rate": 1.6746034850443683e-05,
|
| 925 |
+
"loss": 2.271,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.34070221066319895,
|
| 930 |
+
"grad_norm": 0.42183587732698574,
|
| 931 |
+
"learning_rate": 1.6678693703666327e-05,
|
| 932 |
+
"loss": 2.2766,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3433029908972692,
|
| 937 |
+
"grad_norm": 0.4143980865523107,
|
| 938 |
+
"learning_rate": 1.6610801319957094e-05,
|
| 939 |
+
"loss": 2.2568,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.3459037711313394,
|
| 944 |
+
"grad_norm": 0.42248752750572977,
|
| 945 |
+
"learning_rate": 1.654236330292515e-05,
|
| 946 |
+
"loss": 2.2685,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.34850455136540964,
|
| 951 |
+
"grad_norm": 0.40537525032067523,
|
| 952 |
+
"learning_rate": 1.647338530121441e-05,
|
| 953 |
+
"loss": 2.2612,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.3511053315994798,
|
| 958 |
+
"grad_norm": 0.4027010096639477,
|
| 959 |
+
"learning_rate": 1.6403873008037306e-05,
|
| 960 |
+
"loss": 2.269,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.35370611183355005,
|
| 965 |
+
"grad_norm": 0.41439590187756714,
|
| 966 |
+
"learning_rate": 1.6333832160704895e-05,
|
| 967 |
+
"loss": 2.2619,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.3563068920676203,
|
| 972 |
+
"grad_norm": 0.4024453337904974,
|
| 973 |
+
"learning_rate": 1.6263268540153313e-05,
|
| 974 |
+
"loss": 2.2639,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.3589076723016905,
|
| 979 |
+
"grad_norm": 0.4104644349765972,
|
| 980 |
+
"learning_rate": 1.6192187970466646e-05,
|
| 981 |
+
"loss": 2.2758,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.36150845253576075,
|
| 986 |
+
"grad_norm": 0.5594987533069902,
|
| 987 |
+
"learning_rate": 1.6120596318396214e-05,
|
| 988 |
+
"loss": 2.2736,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3641092327698309,
|
| 993 |
+
"grad_norm": 0.404796094364898,
|
| 994 |
+
"learning_rate": 1.6048499492876378e-05,
|
| 995 |
+
"loss": 2.2959,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.36671001300390116,
|
| 1000 |
+
"grad_norm": 0.41993515236832657,
|
| 1001 |
+
"learning_rate": 1.5975903444536798e-05,
|
| 1002 |
+
"loss": 2.2932,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3693107932379714,
|
| 1007 |
+
"grad_norm": 0.41692831995756224,
|
| 1008 |
+
"learning_rate": 1.5902814165211314e-05,
|
| 1009 |
+
"loss": 2.2739,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.3719115734720416,
|
| 1014 |
+
"grad_norm": 0.4088005848052102,
|
| 1015 |
+
"learning_rate": 1.5829237687443407e-05,
|
| 1016 |
+
"loss": 2.2741,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.37451235370611186,
|
| 1021 |
+
"grad_norm": 0.4348987588147031,
|
| 1022 |
+
"learning_rate": 1.5755180083988256e-05,
|
| 1023 |
+
"loss": 2.299,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.37711313394018203,
|
| 1028 |
+
"grad_norm": 0.3869333371066437,
|
| 1029 |
+
"learning_rate": 1.568064746731156e-05,
|
| 1030 |
+
"loss": 2.2736,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.37971391417425226,
|
| 1035 |
+
"grad_norm": 0.4724079277564694,
|
| 1036 |
+
"learning_rate": 1.5605645989084996e-05,
|
| 1037 |
+
"loss": 2.2574,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.3823146944083225,
|
| 1042 |
+
"grad_norm": 0.42727310538612284,
|
| 1043 |
+
"learning_rate": 1.5530181839678506e-05,
|
| 1044 |
+
"loss": 2.2559,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.38491547464239273,
|
| 1049 |
+
"grad_norm": 0.4149607685779955,
|
| 1050 |
+
"learning_rate": 1.545426124764936e-05,
|
| 1051 |
+
"loss": 2.2747,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.38751625487646296,
|
| 1056 |
+
"grad_norm": 0.42562746149976444,
|
| 1057 |
+
"learning_rate": 1.5377890479228053e-05,
|
| 1058 |
+
"loss": 2.2686,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.39011703511053314,
|
| 1063 |
+
"grad_norm": 0.42242365905288914,
|
| 1064 |
+
"learning_rate": 1.5301075837801145e-05,
|
| 1065 |
+
"loss": 2.262,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.39271781534460337,
|
| 1070 |
+
"grad_norm": 0.4043182375613442,
|
| 1071 |
+
"learning_rate": 1.5223823663390964e-05,
|
| 1072 |
+
"loss": 2.2999,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.3953185955786736,
|
| 1077 |
+
"grad_norm": 0.4014226795792646,
|
| 1078 |
+
"learning_rate": 1.5146140332132359e-05,
|
| 1079 |
+
"loss": 2.2882,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.39791937581274384,
|
| 1084 |
+
"grad_norm": 0.41229108021598954,
|
| 1085 |
+
"learning_rate": 1.50680322557464e-05,
|
| 1086 |
+
"loss": 2.2915,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.40052015604681407,
|
| 1091 |
+
"grad_norm": 0.40410628915367974,
|
| 1092 |
+
"learning_rate": 1.4989505881011201e-05,
|
| 1093 |
+
"loss": 2.2827,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.40312093628088425,
|
| 1098 |
+
"grad_norm": 0.4123788805293143,
|
| 1099 |
+
"learning_rate": 1.4910567689229812e-05,
|
| 1100 |
+
"loss": 2.2758,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4057217165149545,
|
| 1105 |
+
"grad_norm": 0.4127020336028451,
|
| 1106 |
+
"learning_rate": 1.4831224195695275e-05,
|
| 1107 |
+
"loss": 2.2866,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4083224967490247,
|
| 1112 |
+
"grad_norm": 0.4067660692602614,
|
| 1113 |
+
"learning_rate": 1.4751481949152888e-05,
|
| 1114 |
+
"loss": 2.248,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.41092327698309494,
|
| 1119 |
+
"grad_norm": 0.40648002372954717,
|
| 1120 |
+
"learning_rate": 1.4671347531259674e-05,
|
| 1121 |
+
"loss": 2.2856,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4135240572171652,
|
| 1126 |
+
"grad_norm": 0.4009001987782148,
|
| 1127 |
+
"learning_rate": 1.4590827556041158e-05,
|
| 1128 |
+
"loss": 2.2905,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.41612483745123535,
|
| 1133 |
+
"grad_norm": 0.4190576870475519,
|
| 1134 |
+
"learning_rate": 1.450992866934548e-05,
|
| 1135 |
+
"loss": 2.2676,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4187256176853056,
|
| 1140 |
+
"grad_norm": 0.40565344687436516,
|
| 1141 |
+
"learning_rate": 1.442865754829485e-05,
|
| 1142 |
+
"loss": 2.2725,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4213263979193758,
|
| 1147 |
+
"grad_norm": 0.39987691654168583,
|
| 1148 |
+
"learning_rate": 1.4347020900734458e-05,
|
| 1149 |
+
"loss": 2.279,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.42392717815344605,
|
| 1154 |
+
"grad_norm": 0.3849016079800747,
|
| 1155 |
+
"learning_rate": 1.426502546467882e-05,
|
| 1156 |
+
"loss": 2.2549,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4265279583875163,
|
| 1161 |
+
"grad_norm": 0.404786678502953,
|
| 1162 |
+
"learning_rate": 1.4182678007755653e-05,
|
| 1163 |
+
"loss": 2.248,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.42912873862158646,
|
| 1168 |
+
"grad_norm": 0.4167196866729737,
|
| 1169 |
+
"learning_rate": 1.4099985326647285e-05,
|
| 1170 |
+
"loss": 2.273,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4317295188556567,
|
| 1175 |
+
"grad_norm": 0.3915200348060608,
|
| 1176 |
+
"learning_rate": 1.4016954246529697e-05,
|
| 1177 |
+
"loss": 2.246,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.4343302990897269,
|
| 1182 |
+
"grad_norm": 0.40449571430111425,
|
| 1183 |
+
"learning_rate": 1.393359162050918e-05,
|
| 1184 |
+
"loss": 2.2364,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.43693107932379716,
|
| 1189 |
+
"grad_norm": 0.3867658749935242,
|
| 1190 |
+
"learning_rate": 1.384990432905671e-05,
|
| 1191 |
+
"loss": 2.2747,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.43953185955786733,
|
| 1196 |
+
"grad_norm": 0.4015710624164935,
|
| 1197 |
+
"learning_rate": 1.3765899279440064e-05,
|
| 1198 |
+
"loss": 2.2791,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.44213263979193757,
|
| 1203 |
+
"grad_norm": 0.40563402700708584,
|
| 1204 |
+
"learning_rate": 1.3681583405153704e-05,
|
| 1205 |
+
"loss": 2.2609,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.4447334200260078,
|
| 1210 |
+
"grad_norm": 0.4118299528517327,
|
| 1211 |
+
"learning_rate": 1.3596963665346514e-05,
|
| 1212 |
+
"loss": 2.2753,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.44733420026007803,
|
| 1217 |
+
"grad_norm": 0.39884845615523695,
|
| 1218 |
+
"learning_rate": 1.3512047044247423e-05,
|
| 1219 |
+
"loss": 2.2819,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.44993498049414826,
|
| 1224 |
+
"grad_norm": 0.408102295409258,
|
| 1225 |
+
"learning_rate": 1.3426840550588933e-05,
|
| 1226 |
+
"loss": 2.2495,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.45253576072821844,
|
| 1231 |
+
"grad_norm": 0.41987937512067763,
|
| 1232 |
+
"learning_rate": 1.3341351217028669e-05,
|
| 1233 |
+
"loss": 2.2487,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.45513654096228867,
|
| 1238 |
+
"grad_norm": 0.39473938306748274,
|
| 1239 |
+
"learning_rate": 1.3255586099568896e-05,
|
| 1240 |
+
"loss": 2.2784,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4577373211963589,
|
| 1245 |
+
"grad_norm": 0.40572104385822627,
|
| 1246 |
+
"learning_rate": 1.3169552276974162e-05,
|
| 1247 |
+
"loss": 2.2657,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.46033810143042914,
|
| 1252 |
+
"grad_norm": 0.39490922913293697,
|
| 1253 |
+
"learning_rate": 1.3083256850187036e-05,
|
| 1254 |
+
"loss": 2.2657,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.46293888166449937,
|
| 1259 |
+
"grad_norm": 0.4011404454106932,
|
| 1260 |
+
"learning_rate": 1.2996706941742012e-05,
|
| 1261 |
+
"loss": 2.2584,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.46553966189856955,
|
| 1266 |
+
"grad_norm": 0.3934876590340417,
|
| 1267 |
+
"learning_rate": 1.2909909695177647e-05,
|
| 1268 |
+
"loss": 2.2492,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.4681404421326398,
|
| 1273 |
+
"grad_norm": 0.4334665858337218,
|
| 1274 |
+
"learning_rate": 1.2822872274446958e-05,
|
| 1275 |
+
"loss": 2.2716,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.47074122236671,
|
| 1280 |
+
"grad_norm": 0.4009072108584327,
|
| 1281 |
+
"learning_rate": 1.2735601863326124e-05,
|
| 1282 |
+
"loss": 2.2693,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.47334200260078024,
|
| 1287 |
+
"grad_norm": 0.4019876662042798,
|
| 1288 |
+
"learning_rate": 1.2648105664821577e-05,
|
| 1289 |
+
"loss": 2.2499,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.4759427828348505,
|
| 1294 |
+
"grad_norm": 0.394364565475087,
|
| 1295 |
+
"learning_rate": 1.2560390900575472e-05,
|
| 1296 |
+
"loss": 2.2683,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.47854356306892065,
|
| 1301 |
+
"grad_norm": 0.4022232639573475,
|
| 1302 |
+
"learning_rate": 1.2472464810269653e-05,
|
| 1303 |
+
"loss": 2.2354,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.4811443433029909,
|
| 1308 |
+
"grad_norm": 0.4271559987618919,
|
| 1309 |
+
"learning_rate": 1.2384334651028098e-05,
|
| 1310 |
+
"loss": 2.266,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.4837451235370611,
|
| 1315 |
+
"grad_norm": 0.4080839682795773,
|
| 1316 |
+
"learning_rate": 1.2296007696817957e-05,
|
| 1317 |
+
"loss": 2.2641,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.48634590377113135,
|
| 1322 |
+
"grad_norm": 0.4169798937817002,
|
| 1323 |
+
"learning_rate": 1.2207491237849174e-05,
|
| 1324 |
+
"loss": 2.2461,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.4889466840052016,
|
| 1329 |
+
"grad_norm": 0.39043603769224566,
|
| 1330 |
+
"learning_rate": 1.2118792579972773e-05,
|
| 1331 |
+
"loss": 2.2574,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.49154746423927176,
|
| 1336 |
+
"grad_norm": 0.41837982163357523,
|
| 1337 |
+
"learning_rate": 1.2029919044077874e-05,
|
| 1338 |
+
"loss": 2.2826,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.494148244473342,
|
| 1343 |
+
"grad_norm": 0.41194970118458496,
|
| 1344 |
+
"learning_rate": 1.194087796548743e-05,
|
| 1345 |
+
"loss": 2.2769,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.4967490247074122,
|
| 1350 |
+
"grad_norm": 0.4040702919109425,
|
| 1351 |
+
"learning_rate": 1.185167669335281e-05,
|
| 1352 |
+
"loss": 2.2977,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.49934980494148246,
|
| 1357 |
+
"grad_norm": 0.39486895318415,
|
| 1358 |
+
"learning_rate": 1.176232259004722e-05,
|
| 1359 |
+
"loss": 2.2652,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5019505851755527,
|
| 1364 |
+
"grad_norm": 0.4094337034652758,
|
| 1365 |
+
"learning_rate": 1.1672823030558036e-05,
|
| 1366 |
+
"loss": 2.2859,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5045513654096229,
|
| 1371 |
+
"grad_norm": 0.3947877180935096,
|
| 1372 |
+
"learning_rate": 1.15831854018781e-05,
|
| 1373 |
+
"loss": 2.2563,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5071521456436932,
|
| 1378 |
+
"grad_norm": 0.3981699504831132,
|
| 1379 |
+
"learning_rate": 1.1493417102396023e-05,
|
| 1380 |
+
"loss": 2.2586,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5097529258777633,
|
| 1385 |
+
"grad_norm": 0.3997818914286282,
|
| 1386 |
+
"learning_rate": 1.1403525541285536e-05,
|
| 1387 |
+
"loss": 2.2529,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5123537061118335,
|
| 1392 |
+
"grad_norm": 0.4034382921172292,
|
| 1393 |
+
"learning_rate": 1.1313518137893984e-05,
|
| 1394 |
+
"loss": 2.2626,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5149544863459038,
|
| 1399 |
+
"grad_norm": 0.40310724786384267,
|
| 1400 |
+
"learning_rate": 1.122340232112993e-05,
|
| 1401 |
+
"loss": 2.2575,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.517555266579974,
|
| 1406 |
+
"grad_norm": 0.3942328864184433,
|
| 1407 |
+
"learning_rate": 1.1133185528850024e-05,
|
| 1408 |
+
"loss": 2.2651,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5201560468140443,
|
| 1413 |
+
"grad_norm": 0.4173488156503363,
|
| 1414 |
+
"learning_rate": 1.104287520724509e-05,
|
| 1415 |
+
"loss": 2.2544,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5227568270481144,
|
| 1420 |
+
"grad_norm": 0.39388806138277255,
|
| 1421 |
+
"learning_rate": 1.095247881022555e-05,
|
| 1422 |
+
"loss": 2.2937,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5253576072821846,
|
| 1427 |
+
"grad_norm": 0.40035476964034117,
|
| 1428 |
+
"learning_rate": 1.0862003798806195e-05,
|
| 1429 |
+
"loss": 2.2872,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5279583875162549,
|
| 1434 |
+
"grad_norm": 0.4024014335133922,
|
| 1435 |
+
"learning_rate": 1.07714576404904e-05,
|
| 1436 |
+
"loss": 2.2687,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5305591677503251,
|
| 1441 |
+
"grad_norm": 0.38731591485476247,
|
| 1442 |
+
"learning_rate": 1.0680847808653747e-05,
|
| 1443 |
+
"loss": 2.2622,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5331599479843954,
|
| 1448 |
+
"grad_norm": 0.40054108594934995,
|
| 1449 |
+
"learning_rate": 1.0590181781927229e-05,
|
| 1450 |
+
"loss": 2.2669,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5357607282184655,
|
| 1455 |
+
"grad_norm": 0.3846726780523599,
|
| 1456 |
+
"learning_rate": 1.0499467043579974e-05,
|
| 1457 |
+
"loss": 2.2641,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5383615084525357,
|
| 1462 |
+
"grad_norm": 0.39464718229343276,
|
| 1463 |
+
"learning_rate": 1.0408711080901598e-05,
|
| 1464 |
+
"loss": 2.2711,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.540962288686606,
|
| 1469 |
+
"grad_norm": 0.40060852826239435,
|
| 1470 |
+
"learning_rate": 1.0317921384584245e-05,
|
| 1471 |
+
"loss": 2.2589,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5435630689206762,
|
| 1476 |
+
"grad_norm": 0.3864095081088353,
|
| 1477 |
+
"learning_rate": 1.0227105448104308e-05,
|
| 1478 |
+
"loss": 2.2757,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5461638491547465,
|
| 1483 |
+
"grad_norm": 0.40683220249643803,
|
| 1484 |
+
"learning_rate": 1.0136270767103967e-05,
|
| 1485 |
+
"loss": 2.2478,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5487646293888166,
|
| 1490 |
+
"grad_norm": 0.40364574316169327,
|
| 1491 |
+
"learning_rate": 1.0045424838772497e-05,
|
| 1492 |
+
"loss": 2.2457,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5513654096228868,
|
| 1497 |
+
"grad_norm": 0.39909089742018006,
|
| 1498 |
+
"learning_rate": 9.954575161227503e-06,
|
| 1499 |
+
"loss": 2.269,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5539661898569571,
|
| 1504 |
+
"grad_norm": 0.3895157377499068,
|
| 1505 |
+
"learning_rate": 9.863729232896035e-06,
|
| 1506 |
+
"loss": 2.272,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5565669700910273,
|
| 1511 |
+
"grad_norm": 0.39908721011359033,
|
| 1512 |
+
"learning_rate": 9.772894551895692e-06,
|
| 1513 |
+
"loss": 2.2598,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5591677503250976,
|
| 1518 |
+
"grad_norm": 0.4145617786401608,
|
| 1519 |
+
"learning_rate": 9.682078615415755e-06,
|
| 1520 |
+
"loss": 2.256,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5617685305591678,
|
| 1525 |
+
"grad_norm": 0.40100222100669214,
|
| 1526 |
+
"learning_rate": 9.591288919098402e-06,
|
| 1527 |
+
"loss": 2.2577,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5643693107932379,
|
| 1532 |
+
"grad_norm": 0.4061162317910723,
|
| 1533 |
+
"learning_rate": 9.500532956420028e-06,
|
| 1534 |
+
"loss": 2.2444,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5669700910273082,
|
| 1539 |
+
"grad_norm": 0.3979878981899698,
|
| 1540 |
+
"learning_rate": 9.409818218072774e-06,
|
| 1541 |
+
"loss": 2.238,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5695708712613784,
|
| 1546 |
+
"grad_norm": 0.42032443863245716,
|
| 1547 |
+
"learning_rate": 9.319152191346255e-06,
|
| 1548 |
+
"loss": 2.259,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5721716514954487,
|
| 1553 |
+
"grad_norm": 0.3841670620130304,
|
| 1554 |
+
"learning_rate": 9.228542359509604e-06,
|
| 1555 |
+
"loss": 2.2726,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5747724317295189,
|
| 1560 |
+
"grad_norm": 0.3934375851231943,
|
| 1561 |
+
"learning_rate": 9.137996201193807e-06,
|
| 1562 |
+
"loss": 2.2817,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.577373211963589,
|
| 1567 |
+
"grad_norm": 0.43987340826594445,
|
| 1568 |
+
"learning_rate": 9.047521189774456e-06,
|
| 1569 |
+
"loss": 2.266,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5799739921976593,
|
| 1574 |
+
"grad_norm": 0.39405209949111636,
|
| 1575 |
+
"learning_rate": 8.957124792754913e-06,
|
| 1576 |
+
"loss": 2.2622,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5825747724317295,
|
| 1581 |
+
"grad_norm": 0.39432828211359744,
|
| 1582 |
+
"learning_rate": 8.866814471149977e-06,
|
| 1583 |
+
"loss": 2.2647,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5851755526657998,
|
| 1588 |
+
"grad_norm": 0.4151146193247924,
|
| 1589 |
+
"learning_rate": 8.776597678870072e-06,
|
| 1590 |
+
"loss": 2.2811,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.58777633289987,
|
| 1595 |
+
"grad_norm": 0.3999401443902343,
|
| 1596 |
+
"learning_rate": 8.686481862106019e-06,
|
| 1597 |
+
"loss": 2.2585,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.5903771131339401,
|
| 1602 |
+
"grad_norm": 0.39080240935002436,
|
| 1603 |
+
"learning_rate": 8.596474458714466e-06,
|
| 1604 |
+
"loss": 2.2608,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.5929778933680104,
|
| 1609 |
+
"grad_norm": 0.3968422406166828,
|
| 1610 |
+
"learning_rate": 8.506582897603979e-06,
|
| 1611 |
+
"loss": 2.2739,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.5955786736020806,
|
| 1616 |
+
"grad_norm": 0.407166366301263,
|
| 1617 |
+
"learning_rate": 8.416814598121901e-06,
|
| 1618 |
+
"loss": 2.2555,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.5981794538361509,
|
| 1623 |
+
"grad_norm": 0.3933361067533457,
|
| 1624 |
+
"learning_rate": 8.327176969441968e-06,
|
| 1625 |
+
"loss": 2.2587,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6007802340702211,
|
| 1630 |
+
"grad_norm": 0.3885172716443014,
|
| 1631 |
+
"learning_rate": 8.237677409952784e-06,
|
| 1632 |
+
"loss": 2.2553,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6033810143042913,
|
| 1637 |
+
"grad_norm": 0.39355008070699726,
|
| 1638 |
+
"learning_rate": 8.148323306647195e-06,
|
| 1639 |
+
"loss": 2.2766,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6059817945383615,
|
| 1644 |
+
"grad_norm": 0.3917151824209379,
|
| 1645 |
+
"learning_rate": 8.059122034512577e-06,
|
| 1646 |
+
"loss": 2.2569,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6085825747724317,
|
| 1651 |
+
"grad_norm": 0.40851029009076756,
|
| 1652 |
+
"learning_rate": 7.970080955922132e-06,
|
| 1653 |
+
"loss": 2.258,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.611183355006502,
|
| 1658 |
+
"grad_norm": 0.39444434752805385,
|
| 1659 |
+
"learning_rate": 7.88120742002723e-06,
|
| 1660 |
+
"loss": 2.2924,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6137841352405722,
|
| 1665 |
+
"grad_norm": 0.4245658805720807,
|
| 1666 |
+
"learning_rate": 7.792508762150833e-06,
|
| 1667 |
+
"loss": 2.2634,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6163849154746424,
|
| 1672 |
+
"grad_norm": 0.41425927958212905,
|
| 1673 |
+
"learning_rate": 7.703992303182044e-06,
|
| 1674 |
+
"loss": 2.2554,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6189856957087126,
|
| 1679 |
+
"grad_norm": 0.4081554322329225,
|
| 1680 |
+
"learning_rate": 7.615665348971903e-06,
|
| 1681 |
+
"loss": 2.2613,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6215864759427828,
|
| 1686 |
+
"grad_norm": 0.41034486699879885,
|
| 1687 |
+
"learning_rate": 7.527535189730351e-06,
|
| 1688 |
+
"loss": 2.2619,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6241872561768531,
|
| 1693 |
+
"grad_norm": 0.42456676190279796,
|
| 1694 |
+
"learning_rate": 7.4396090994245295e-06,
|
| 1695 |
+
"loss": 2.247,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6267880364109233,
|
| 1700 |
+
"grad_norm": 0.3819217051140379,
|
| 1701 |
+
"learning_rate": 7.351894335178424e-06,
|
| 1702 |
+
"loss": 2.2696,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6293888166449935,
|
| 1707 |
+
"grad_norm": 0.4034781104713241,
|
| 1708 |
+
"learning_rate": 7.264398136673877e-06,
|
| 1709 |
+
"loss": 2.2811,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6319895968790638,
|
| 1714 |
+
"grad_norm": 0.38862943048622406,
|
| 1715 |
+
"learning_rate": 7.1771277255530456e-06,
|
| 1716 |
+
"loss": 2.2766,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6345903771131339,
|
| 1721 |
+
"grad_norm": 0.40127806453021636,
|
| 1722 |
+
"learning_rate": 7.090090304822356e-06,
|
| 1723 |
+
"loss": 2.2697,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6371911573472041,
|
| 1728 |
+
"grad_norm": 0.41972950579228285,
|
| 1729 |
+
"learning_rate": 7.003293058257991e-06,
|
| 1730 |
+
"loss": 2.2652,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6397919375812744,
|
| 1735 |
+
"grad_norm": 0.427234544107129,
|
| 1736 |
+
"learning_rate": 6.916743149812968e-06,
|
| 1737 |
+
"loss": 2.2669,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6423927178153446,
|
| 1742 |
+
"grad_norm": 0.39336959620711115,
|
| 1743 |
+
"learning_rate": 6.8304477230258416e-06,
|
| 1744 |
+
"loss": 2.2328,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6449934980494149,
|
| 1749 |
+
"grad_norm": 0.41464412249859484,
|
| 1750 |
+
"learning_rate": 6.744413900431109e-06,
|
| 1751 |
+
"loss": 2.269,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.647594278283485,
|
| 1756 |
+
"grad_norm": 0.407611910961648,
|
| 1757 |
+
"learning_rate": 6.658648782971335e-06,
|
| 1758 |
+
"loss": 2.2721,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6501950585175552,
|
| 1763 |
+
"grad_norm": 0.49124182873138217,
|
| 1764 |
+
"learning_rate": 6.573159449411071e-06,
|
| 1765 |
+
"loss": 2.2719,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6527958387516255,
|
| 1770 |
+
"grad_norm": 0.3903162593306489,
|
| 1771 |
+
"learning_rate": 6.487952955752581e-06,
|
| 1772 |
+
"loss": 2.2883,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6553966189856957,
|
| 1777 |
+
"grad_norm": 0.3831301362957459,
|
| 1778 |
+
"learning_rate": 6.403036334653487e-06,
|
| 1779 |
+
"loss": 2.2658,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.657997399219766,
|
| 1784 |
+
"grad_norm": 0.38558455885900716,
|
| 1785 |
+
"learning_rate": 6.3184165948463e-06,
|
| 1786 |
+
"loss": 2.2744,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6605981794538361,
|
| 1791 |
+
"grad_norm": 0.4017103958771505,
|
| 1792 |
+
"learning_rate": 6.234100720559939e-06,
|
| 1793 |
+
"loss": 2.2511,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6631989596879063,
|
| 1798 |
+
"grad_norm": 0.388129934280965,
|
| 1799 |
+
"learning_rate": 6.150095670943293e-06,
|
| 1800 |
+
"loss": 2.2571,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6657997399219766,
|
| 1805 |
+
"grad_norm": 0.40380196802049817,
|
| 1806 |
+
"learning_rate": 6.066408379490826e-06,
|
| 1807 |
+
"loss": 2.2638,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6684005201560468,
|
| 1812 |
+
"grad_norm": 0.3884972325810741,
|
| 1813 |
+
"learning_rate": 5.983045753470308e-06,
|
| 1814 |
+
"loss": 2.2551,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6710013003901171,
|
| 1819 |
+
"grad_norm": 0.4395057468261991,
|
| 1820 |
+
"learning_rate": 5.9000146733527185e-06,
|
| 1821 |
+
"loss": 2.2714,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6736020806241872,
|
| 1826 |
+
"grad_norm": 0.39807884730304743,
|
| 1827 |
+
"learning_rate": 5.8173219922443516e-06,
|
| 1828 |
+
"loss": 2.2598,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6762028608582574,
|
| 1833 |
+
"grad_norm": 0.385727954763992,
|
| 1834 |
+
"learning_rate": 5.734974535321183e-06,
|
| 1835 |
+
"loss": 2.27,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6788036410923277,
|
| 1840 |
+
"grad_norm": 0.3952142446123414,
|
| 1841 |
+
"learning_rate": 5.652979099265546e-06,
|
| 1842 |
+
"loss": 2.2938,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6814044213263979,
|
| 1847 |
+
"grad_norm": 0.38603580794024683,
|
| 1848 |
+
"learning_rate": 5.571342451705151e-06,
|
| 1849 |
+
"loss": 2.2452,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6840052015604682,
|
| 1854 |
+
"grad_norm": 0.390022304958138,
|
| 1855 |
+
"learning_rate": 5.490071330654523e-06,
|
| 1856 |
+
"loss": 2.2627,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6866059817945384,
|
| 1861 |
+
"grad_norm": 0.38894369305561394,
|
| 1862 |
+
"learning_rate": 5.409172443958844e-06,
|
| 1863 |
+
"loss": 2.2635,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6892067620286085,
|
| 1868 |
+
"grad_norm": 0.39275649980316757,
|
| 1869 |
+
"learning_rate": 5.328652468740328e-06,
|
| 1870 |
+
"loss": 2.2498,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.6918075422626788,
|
| 1875 |
+
"grad_norm": 0.3962525545592007,
|
| 1876 |
+
"learning_rate": 5.248518050847113e-06,
|
| 1877 |
+
"loss": 2.2468,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.694408322496749,
|
| 1882 |
+
"grad_norm": 0.3903498521635712,
|
| 1883 |
+
"learning_rate": 5.168775804304727e-06,
|
| 1884 |
+
"loss": 2.2747,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6970091027308193,
|
| 1889 |
+
"grad_norm": 0.38636928696844103,
|
| 1890 |
+
"learning_rate": 5.089432310770195e-06,
|
| 1891 |
+
"loss": 2.2478,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.6996098829648895,
|
| 1896 |
+
"grad_norm": 0.3774623370136679,
|
| 1897 |
+
"learning_rate": 5.0104941189888e-06,
|
| 1898 |
+
"loss": 2.2546,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7022106631989596,
|
| 1903 |
+
"grad_norm": 0.39156537505231703,
|
| 1904 |
+
"learning_rate": 4.931967744253601e-06,
|
| 1905 |
+
"loss": 2.257,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7048114434330299,
|
| 1910 |
+
"grad_norm": 0.3909178221858017,
|
| 1911 |
+
"learning_rate": 4.853859667867641e-06,
|
| 1912 |
+
"loss": 2.2569,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7074122236671001,
|
| 1917 |
+
"grad_norm": 0.37588492290442793,
|
| 1918 |
+
"learning_rate": 4.776176336609035e-06,
|
| 1919 |
+
"loss": 2.2371,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7100130039011704,
|
| 1924 |
+
"grad_norm": 0.40009668516680985,
|
| 1925 |
+
"learning_rate": 4.698924162198858e-06,
|
| 1926 |
+
"loss": 2.2687,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7126137841352406,
|
| 1931 |
+
"grad_norm": 0.38553642206468475,
|
| 1932 |
+
"learning_rate": 4.622109520771948e-06,
|
| 1933 |
+
"loss": 2.2775,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7152145643693107,
|
| 1938 |
+
"grad_norm": 0.39325643204672056,
|
| 1939 |
+
"learning_rate": 4.545738752350643e-06,
|
| 1940 |
+
"loss": 2.2529,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.717815344603381,
|
| 1945 |
+
"grad_norm": 0.3860361299092455,
|
| 1946 |
+
"learning_rate": 4.469818160321495e-06,
|
| 1947 |
+
"loss": 2.2448,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7204161248374512,
|
| 1952 |
+
"grad_norm": 0.3801889806907439,
|
| 1953 |
+
"learning_rate": 4.394354010915006e-06,
|
| 1954 |
+
"loss": 2.2583,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7230169050715215,
|
| 1959 |
+
"grad_norm": 0.3939241087166142,
|
| 1960 |
+
"learning_rate": 4.319352532688444e-06,
|
| 1961 |
+
"loss": 2.289,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7256176853055917,
|
| 1966 |
+
"grad_norm": 0.38993321484677773,
|
| 1967 |
+
"learning_rate": 4.244819916011747e-06,
|
| 1968 |
+
"loss": 2.2702,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7282184655396619,
|
| 1973 |
+
"grad_norm": 0.388552488346714,
|
| 1974 |
+
"learning_rate": 4.170762312556598e-06,
|
| 1975 |
+
"loss": 2.2752,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7308192457737321,
|
| 1980 |
+
"grad_norm": 0.4207163454190188,
|
| 1981 |
+
"learning_rate": 4.097185834788687e-06,
|
| 1982 |
+
"loss": 2.2626,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7334200260078023,
|
| 1987 |
+
"grad_norm": 0.3951814890975275,
|
| 1988 |
+
"learning_rate": 4.024096555463208e-06,
|
| 1989 |
+
"loss": 2.2683,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7360208062418726,
|
| 1994 |
+
"grad_norm": 0.38507477260126755,
|
| 1995 |
+
"learning_rate": 3.9515005071236274e-06,
|
| 1996 |
+
"loss": 2.2838,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7386215864759428,
|
| 2001 |
+
"grad_norm": 0.38407848712158443,
|
| 2002 |
+
"learning_rate": 3.879403681603789e-06,
|
| 2003 |
+
"loss": 2.2474,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.741222366710013,
|
| 2008 |
+
"grad_norm": 0.3854544788178802,
|
| 2009 |
+
"learning_rate": 3.807812029533362e-06,
|
| 2010 |
+
"loss": 2.2765,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7438231469440832,
|
| 2015 |
+
"grad_norm": 0.3863389213391131,
|
| 2016 |
+
"learning_rate": 3.736731459846693e-06,
|
| 2017 |
+
"loss": 2.2882,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7464239271781534,
|
| 2022 |
+
"grad_norm": 0.40550203977347526,
|
| 2023 |
+
"learning_rate": 3.6661678392951085e-06,
|
| 2024 |
+
"loss": 2.2728,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7490247074122237,
|
| 2029 |
+
"grad_norm": 0.3875431885330012,
|
| 2030 |
+
"learning_rate": 3.596126991962696e-06,
|
| 2031 |
+
"loss": 2.2675,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7516254876462939,
|
| 2036 |
+
"grad_norm": 0.38267682697036776,
|
| 2037 |
+
"learning_rate": 3.526614698785593e-06,
|
| 2038 |
+
"loss": 2.2663,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7542262678803641,
|
| 2043 |
+
"grad_norm": 0.3829405792323349,
|
| 2044 |
+
"learning_rate": 3.4576366970748533e-06,
|
| 2045 |
+
"loss": 2.2579,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7568270481144344,
|
| 2050 |
+
"grad_norm": 0.3788914559018415,
|
| 2051 |
+
"learning_rate": 3.389198680042907e-06,
|
| 2052 |
+
"loss": 2.215,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7594278283485045,
|
| 2057 |
+
"grad_norm": 0.38640140872036,
|
| 2058 |
+
"learning_rate": 3.321306296333673e-06,
|
| 2059 |
+
"loss": 2.2661,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7620286085825748,
|
| 2064 |
+
"grad_norm": 0.37331885863291814,
|
| 2065 |
+
"learning_rate": 3.2539651495563196e-06,
|
| 2066 |
+
"loss": 2.2257,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.764629388816645,
|
| 2071 |
+
"grad_norm": 0.3896698522926214,
|
| 2072 |
+
"learning_rate": 3.1871807978227755e-06,
|
| 2073 |
+
"loss": 2.2649,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7672301690507152,
|
| 2078 |
+
"grad_norm": 0.3945500552862151,
|
| 2079 |
+
"learning_rate": 3.1209587532889763e-06,
|
| 2080 |
+
"loss": 2.2749,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7698309492847855,
|
| 2085 |
+
"grad_norm": 0.3887167709088591,
|
| 2086 |
+
"learning_rate": 3.0553044816999133e-06,
|
| 2087 |
+
"loss": 2.251,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7724317295188556,
|
| 2092 |
+
"grad_norm": 0.3953562634326499,
|
| 2093 |
+
"learning_rate": 2.9902234019385056e-06,
|
| 2094 |
+
"loss": 2.224,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7750325097529259,
|
| 2099 |
+
"grad_norm": 0.3929809796676909,
|
| 2100 |
+
"learning_rate": 2.9257208855783457e-06,
|
| 2101 |
+
"loss": 2.2748,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7776332899869961,
|
| 2106 |
+
"grad_norm": 0.38763794424838,
|
| 2107 |
+
"learning_rate": 2.861802256440348e-06,
|
| 2108 |
+
"loss": 2.2819,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7802340702210663,
|
| 2113 |
+
"grad_norm": 0.38081085068273157,
|
| 2114 |
+
"learning_rate": 2.7984727901533373e-06,
|
| 2115 |
+
"loss": 2.2856,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.7828348504551366,
|
| 2120 |
+
"grad_norm": 0.3892127183637425,
|
| 2121 |
+
"learning_rate": 2.735737713718616e-06,
|
| 2122 |
+
"loss": 2.2488,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.7854356306892067,
|
| 2127 |
+
"grad_norm": 0.3802332515715518,
|
| 2128 |
+
"learning_rate": 2.673602205078547e-06,
|
| 2129 |
+
"loss": 2.2511,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.788036410923277,
|
| 2134 |
+
"grad_norm": 0.38733340356596635,
|
| 2135 |
+
"learning_rate": 2.612071392689184e-06,
|
| 2136 |
+
"loss": 2.2591,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.7906371911573472,
|
| 2141 |
+
"grad_norm": 0.39010995545634564,
|
| 2142 |
+
"learning_rate": 2.5511503550969796e-06,
|
| 2143 |
+
"loss": 2.2601,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.7932379713914174,
|
| 2148 |
+
"grad_norm": 0.38609021809835253,
|
| 2149 |
+
"learning_rate": 2.49084412051963e-06,
|
| 2150 |
+
"loss": 2.2736,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7958387516254877,
|
| 2155 |
+
"grad_norm": 0.3776811943289756,
|
| 2156 |
+
"learning_rate": 2.431157666431052e-06,
|
| 2157 |
+
"loss": 2.2509,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.7984395318595578,
|
| 2162 |
+
"grad_norm": 0.40263080301881393,
|
| 2163 |
+
"learning_rate": 2.3720959191505676e-06,
|
| 2164 |
+
"loss": 2.2412,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8010403120936281,
|
| 2169 |
+
"grad_norm": 0.3775899001896076,
|
| 2170 |
+
"learning_rate": 2.313663753436296e-06,
|
| 2171 |
+
"loss": 2.2307,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8036410923276983,
|
| 2176 |
+
"grad_norm": 0.38311624653572995,
|
| 2177 |
+
"learning_rate": 2.2558659920828095e-06,
|
| 2178 |
+
"loss": 2.2743,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8062418725617685,
|
| 2183 |
+
"grad_norm": 0.3907924186130041,
|
| 2184 |
+
"learning_rate": 2.1987074055230775e-06,
|
| 2185 |
+
"loss": 2.2582,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8088426527958388,
|
| 2190 |
+
"grad_norm": 0.401680147882502,
|
| 2191 |
+
"learning_rate": 2.142192711434733e-06,
|
| 2192 |
+
"loss": 2.2574,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.811443433029909,
|
| 2197 |
+
"grad_norm": 0.39498645257009396,
|
| 2198 |
+
"learning_rate": 2.0863265743506842e-06,
|
| 2199 |
+
"loss": 2.2794,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8140442132639792,
|
| 2204 |
+
"grad_norm": 0.3940744787681164,
|
| 2205 |
+
"learning_rate": 2.0311136052741274e-06,
|
| 2206 |
+
"loss": 2.2686,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8166449934980494,
|
| 2211 |
+
"grad_norm": 0.36893837449198125,
|
| 2212 |
+
"learning_rate": 1.97655836129796e-06,
|
| 2213 |
+
"loss": 2.2685,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8192457737321196,
|
| 2218 |
+
"grad_norm": 0.38517797209515453,
|
| 2219 |
+
"learning_rate": 1.9226653452286703e-06,
|
| 2220 |
+
"loss": 2.286,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8218465539661899,
|
| 2225 |
+
"grad_norm": 0.5482631631538819,
|
| 2226 |
+
"learning_rate": 1.8694390052146737e-06,
|
| 2227 |
+
"loss": 2.2525,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8244473342002601,
|
| 2232 |
+
"grad_norm": 0.37615391003914245,
|
| 2233 |
+
"learning_rate": 1.8168837343791878e-06,
|
| 2234 |
+
"loss": 2.2409,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8270481144343304,
|
| 2239 |
+
"grad_norm": 0.3752132126851401,
|
| 2240 |
+
"learning_rate": 1.7650038704576323e-06,
|
| 2241 |
+
"loss": 2.2862,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8296488946684005,
|
| 2246 |
+
"grad_norm": 0.37537152589883754,
|
| 2247 |
+
"learning_rate": 1.713803695439613e-06,
|
| 2248 |
+
"loss": 2.255,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8322496749024707,
|
| 2253 |
+
"grad_norm": 0.387157412230961,
|
| 2254 |
+
"learning_rate": 1.6632874352154982e-06,
|
| 2255 |
+
"loss": 2.2603,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.834850455136541,
|
| 2260 |
+
"grad_norm": 0.3731805854354132,
|
| 2261 |
+
"learning_rate": 1.6134592592276255e-06,
|
| 2262 |
+
"loss": 2.2483,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8374512353706112,
|
| 2267 |
+
"grad_norm": 0.38456257085878814,
|
| 2268 |
+
"learning_rate": 1.5643232801261731e-06,
|
| 2269 |
+
"loss": 2.2357,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8400520156046815,
|
| 2274 |
+
"grad_norm": 0.38555465933139627,
|
| 2275 |
+
"learning_rate": 1.5158835534297167e-06,
|
| 2276 |
+
"loss": 2.244,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8426527958387516,
|
| 2281 |
+
"grad_norm": 0.37578408042042205,
|
| 2282 |
+
"learning_rate": 1.4681440771904953e-06,
|
| 2283 |
+
"loss": 2.2408,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8452535760728218,
|
| 2288 |
+
"grad_norm": 0.3778467523940946,
|
| 2289 |
+
"learning_rate": 1.4211087916644307e-06,
|
| 2290 |
+
"loss": 2.264,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8478543563068921,
|
| 2295 |
+
"grad_norm": 0.4002340997364397,
|
| 2296 |
+
"learning_rate": 1.3747815789859087e-06,
|
| 2297 |
+
"loss": 2.2357,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8504551365409623,
|
| 2302 |
+
"grad_norm": 0.37692759499605927,
|
| 2303 |
+
"learning_rate": 1.3291662628473634e-06,
|
| 2304 |
+
"loss": 2.2451,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8530559167750326,
|
| 2309 |
+
"grad_norm": 0.38048563348857367,
|
| 2310 |
+
"learning_rate": 1.2842666081836797e-06,
|
| 2311 |
+
"loss": 2.2513,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8556566970091027,
|
| 2316 |
+
"grad_norm": 0.3720985976854136,
|
| 2317 |
+
"learning_rate": 1.240086320861451e-06,
|
| 2318 |
+
"loss": 2.2293,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8582574772431729,
|
| 2323 |
+
"grad_norm": 0.39771984961162216,
|
| 2324 |
+
"learning_rate": 1.1966290473731067e-06,
|
| 2325 |
+
"loss": 2.2587,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8608582574772432,
|
| 2330 |
+
"grad_norm": 0.37533947863605094,
|
| 2331 |
+
"learning_rate": 1.1538983745359467e-06,
|
| 2332 |
+
"loss": 2.2349,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8634590377113134,
|
| 2337 |
+
"grad_norm": 0.37177641638423103,
|
| 2338 |
+
"learning_rate": 1.111897829196088e-06,
|
| 2339 |
+
"loss": 2.2656,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8660598179453837,
|
| 2344 |
+
"grad_norm": 0.3779022618811346,
|
| 2345 |
+
"learning_rate": 1.0706308779373909e-06,
|
| 2346 |
+
"loss": 2.2441,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8686605981794538,
|
| 2351 |
+
"grad_norm": 0.38902026349109015,
|
| 2352 |
+
"learning_rate": 1.0301009267953145e-06,
|
| 2353 |
+
"loss": 2.2508,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.871261378413524,
|
| 2358 |
+
"grad_norm": 0.3916017804328827,
|
| 2359 |
+
"learning_rate": 9.903113209758098e-07,
|
| 2360 |
+
"loss": 2.2595,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8738621586475943,
|
| 2365 |
+
"grad_norm": 0.37948973305758477,
|
| 2366 |
+
"learning_rate": 9.512653445792108e-07,
|
| 2367 |
+
"loss": 2.2229,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.8764629388816645,
|
| 2372 |
+
"grad_norm": 0.3764922488234604,
|
| 2373 |
+
"learning_rate": 9.129662203291789e-07,
|
| 2374 |
+
"loss": 2.2661,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.8790637191157347,
|
| 2379 |
+
"grad_norm": 0.38226936436167364,
|
| 2380 |
+
"learning_rate": 8.754171093067132e-07,
|
| 2381 |
+
"loss": 2.2472,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.881664499349805,
|
| 2386 |
+
"grad_norm": 0.41020670987057306,
|
| 2387 |
+
"learning_rate": 8.386211106892384e-07,
|
| 2388 |
+
"loss": 2.2853,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8842652795838751,
|
| 2393 |
+
"grad_norm": 0.37656002603908223,
|
| 2394 |
+
"learning_rate": 8.025812614948114e-07,
|
| 2395 |
+
"loss": 2.2544,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.8868660598179454,
|
| 2400 |
+
"grad_norm": 0.37970213007035025,
|
| 2401 |
+
"learning_rate": 7.673005363314578e-07,
|
| 2402 |
+
"loss": 2.2726,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.8894668400520156,
|
| 2407 |
+
"grad_norm": 0.39605247830780005,
|
| 2408 |
+
"learning_rate": 7.32781847151659e-07,
|
| 2409 |
+
"loss": 2.2547,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.8920676202860858,
|
| 2414 |
+
"grad_norm": 0.39827415843185704,
|
| 2415 |
+
"learning_rate": 6.990280430120055e-07,
|
| 2416 |
+
"loss": 2.2609,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.8946684005201561,
|
| 2421 |
+
"grad_norm": 0.3803351698581214,
|
| 2422 |
+
"learning_rate": 6.660419098380488e-07,
|
| 2423 |
+
"loss": 2.2609,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.8972691807542262,
|
| 2428 |
+
"grad_norm": 0.3816389494105673,
|
| 2429 |
+
"learning_rate": 6.33826170194356e-07,
|
| 2430 |
+
"loss": 2.257,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.8998699609882965,
|
| 2435 |
+
"grad_norm": 0.38275671555719004,
|
| 2436 |
+
"learning_rate": 6.023834830598041e-07,
|
| 2437 |
+
"loss": 2.2707,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.9024707412223667,
|
| 2442 |
+
"grad_norm": 0.3855930290734069,
|
| 2443 |
+
"learning_rate": 5.717164436081146e-07,
|
| 2444 |
+
"loss": 2.2667,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9050715214564369,
|
| 2449 |
+
"grad_norm": 0.3835325626182195,
|
| 2450 |
+
"learning_rate": 5.418275829936537e-07,
|
| 2451 |
+
"loss": 2.2474,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9076723016905072,
|
| 2456 |
+
"grad_norm": 0.38064681648057413,
|
| 2457 |
+
"learning_rate": 5.127193681425235e-07,
|
| 2458 |
+
"loss": 2.2553,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9102730819245773,
|
| 2463 |
+
"grad_norm": 0.37730514236377943,
|
| 2464 |
+
"learning_rate": 4.843942015489489e-07,
|
| 2465 |
+
"loss": 2.2792,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9128738621586476,
|
| 2470 |
+
"grad_norm": 0.37288039162884873,
|
| 2471 |
+
"learning_rate": 4.5685442107698096e-07,
|
| 2472 |
+
"loss": 2.269,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9154746423927178,
|
| 2477 |
+
"grad_norm": 0.3827871216053965,
|
| 2478 |
+
"learning_rate": 4.301022997675419e-07,
|
| 2479 |
+
"loss": 2.2425,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.918075422626788,
|
| 2484 |
+
"grad_norm": 0.39390214969444604,
|
| 2485 |
+
"learning_rate": 4.041400456508149e-07,
|
| 2486 |
+
"loss": 2.2435,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9206762028608583,
|
| 2491 |
+
"grad_norm": 0.3856070548683713,
|
| 2492 |
+
"learning_rate": 3.7896980156399533e-07,
|
| 2493 |
+
"loss": 2.2742,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9232769830949284,
|
| 2498 |
+
"grad_norm": 0.386403670459497,
|
| 2499 |
+
"learning_rate": 3.5459364497443696e-07,
|
| 2500 |
+
"loss": 2.273,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9258777633289987,
|
| 2505 |
+
"grad_norm": 0.37307256751518914,
|
| 2506 |
+
"learning_rate": 3.3101358780817996e-07,
|
| 2507 |
+
"loss": 2.24,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9284785435630689,
|
| 2512 |
+
"grad_norm": 0.38977960438669146,
|
| 2513 |
+
"learning_rate": 3.0823157628389167e-07,
|
| 2514 |
+
"loss": 2.2373,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9310793237971391,
|
| 2519 |
+
"grad_norm": 0.3812203968889763,
|
| 2520 |
+
"learning_rate": 2.8624949075223505e-07,
|
| 2521 |
+
"loss": 2.2315,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9336801040312094,
|
| 2526 |
+
"grad_norm": 0.38182681988343575,
|
| 2527 |
+
"learning_rate": 2.6506914554067066e-07,
|
| 2528 |
+
"loss": 2.2427,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9362808842652796,
|
| 2533 |
+
"grad_norm": 0.3880543691938056,
|
| 2534 |
+
"learning_rate": 2.446922888037073e-07,
|
| 2535 |
+
"loss": 2.2487,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9388816644993498,
|
| 2540 |
+
"grad_norm": 0.3877865271279748,
|
| 2541 |
+
"learning_rate": 2.2512060237861455e-07,
|
| 2542 |
+
"loss": 2.2632,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.94148244473342,
|
| 2547 |
+
"grad_norm": 0.3790687993984035,
|
| 2548 |
+
"learning_rate": 2.063557016466111e-07,
|
| 2549 |
+
"loss": 2.2485,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9440832249674902,
|
| 2554 |
+
"grad_norm": 0.3828629740071399,
|
| 2555 |
+
"learning_rate": 1.8839913539953957e-07,
|
| 2556 |
+
"loss": 2.2632,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9466840052015605,
|
| 2561 |
+
"grad_norm": 0.3924878743310562,
|
| 2562 |
+
"learning_rate": 1.7125238571202518e-07,
|
| 2563 |
+
"loss": 2.2516,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9492847854356307,
|
| 2568 |
+
"grad_norm": 0.3714180704191971,
|
| 2569 |
+
"learning_rate": 1.5491686781915816e-07,
|
| 2570 |
+
"loss": 2.2192,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.951885565669701,
|
| 2575 |
+
"grad_norm": 0.3775089201491991,
|
| 2576 |
+
"learning_rate": 1.3939392999968493e-07,
|
| 2577 |
+
"loss": 2.2587,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9544863459037711,
|
| 2582 |
+
"grad_norm": 0.3781306612431073,
|
| 2583 |
+
"learning_rate": 1.2468485346472159e-07,
|
| 2584 |
+
"loss": 2.2596,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9570871261378413,
|
| 2589 |
+
"grad_norm": 0.39412747817390337,
|
| 2590 |
+
"learning_rate": 1.1079085225200958e-07,
|
| 2591 |
+
"loss": 2.26,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9596879063719116,
|
| 2596 |
+
"grad_norm": 0.36973691286896965,
|
| 2597 |
+
"learning_rate": 9.771307312571254e-08,
|
| 2598 |
+
"loss": 2.2507,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9622886866059818,
|
| 2603 |
+
"grad_norm": 0.39243425581840186,
|
| 2604 |
+
"learning_rate": 8.545259548176643e-08,
|
| 2605 |
+
"loss": 2.2784,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9648894668400521,
|
| 2610 |
+
"grad_norm": 0.3797955851558652,
|
| 2611 |
+
"learning_rate": 7.40104312587897e-08,
|
| 2612 |
+
"loss": 2.2475,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.9674902470741222,
|
| 2617 |
+
"grad_norm": 0.3841869517905489,
|
| 2618 |
+
"learning_rate": 6.338752485456124e-08,
|
| 2619 |
+
"loss": 2.2599,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9700910273081924,
|
| 2624 |
+
"grad_norm": 0.39253958783577436,
|
| 2625 |
+
"learning_rate": 5.3584753048073756e-08,
|
| 2626 |
+
"loss": 2.24,
|
| 2627 |
+
"step": 1865
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.9726918075422627,
|
| 2631 |
+
"grad_norm": 0.3775787184357385,
|
| 2632 |
+
"learning_rate": 4.460292492716512e-08,
|
| 2633 |
+
"loss": 2.2682,
|
| 2634 |
+
"step": 1870
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.9752925877763329,
|
| 2638 |
+
"grad_norm": 0.39178364824191303,
|
| 2639 |
+
"learning_rate": 3.644278182174166e-08,
|
| 2640 |
+
"loss": 2.2534,
|
| 2641 |
+
"step": 1875
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.9778933680104032,
|
| 2645 |
+
"grad_norm": 0.3810604372347185,
|
| 2646 |
+
"learning_rate": 2.9104997242590528e-08,
|
| 2647 |
+
"loss": 2.2877,
|
| 2648 |
+
"step": 1880
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.9804941482444733,
|
| 2652 |
+
"grad_norm": 0.37376356659580745,
|
| 2653 |
+
"learning_rate": 2.259017682578857e-08,
|
| 2654 |
+
"loss": 2.2899,
|
| 2655 |
+
"step": 1885
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.9830949284785435,
|
| 2659 |
+
"grad_norm": 0.3848779393458631,
|
| 2660 |
+
"learning_rate": 1.6898858282716757e-08,
|
| 2661 |
+
"loss": 2.2361,
|
| 2662 |
+
"step": 1890
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.9856957087126138,
|
| 2666 |
+
"grad_norm": 0.38035967847740393,
|
| 2667 |
+
"learning_rate": 1.2031511355682369e-08,
|
| 2668 |
+
"loss": 2.261,
|
| 2669 |
+
"step": 1895
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.988296488946684,
|
| 2673 |
+
"grad_norm": 0.38901882934105386,
|
| 2674 |
+
"learning_rate": 7.988537779138883e-09,
|
| 2675 |
+
"loss": 2.2654,
|
| 2676 |
+
"step": 1900
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.9908972691807543,
|
| 2680 |
+
"grad_norm": 0.3805341021283482,
|
| 2681 |
+
"learning_rate": 4.770271246539171e-09,
|
| 2682 |
+
"loss": 2.2749,
|
| 2683 |
+
"step": 1905
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.9934980494148244,
|
| 2687 |
+
"grad_norm": 0.3753123691502261,
|
| 2688 |
+
"learning_rate": 2.3769773827864163e-09,
|
| 2689 |
+
"loss": 2.2628,
|
| 2690 |
+
"step": 1910
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.9960988296488946,
|
| 2694 |
+
"grad_norm": 0.38355233778538544,
|
| 2695 |
+
"learning_rate": 8.088537223116533e-10,
|
| 2696 |
+
"loss": 2.2619,
|
| 2697 |
+
"step": 1915
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.9986996098829649,
|
| 2701 |
+
"grad_norm": 0.37535354680698263,
|
| 2702 |
+
"learning_rate": 6.602969277236248e-11,
|
| 2703 |
+
"loss": 2.2856,
|
| 2704 |
+
"step": 1920
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.999739921976593,
|
| 2708 |
+
"eval_loss": 2.262159824371338,
|
| 2709 |
+
"eval_runtime": 106.1225,
|
| 2710 |
+
"eval_samples_per_second": 191.269,
|
| 2711 |
+
"eval_steps_per_second": 2.997,
|
| 2712 |
+
"step": 1922
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.999739921976593,
|
| 2716 |
+
"step": 1922,
|
| 2717 |
+
"total_flos": 1058556698689536.0,
|
| 2718 |
+
"train_loss": 2.26795463333765,
|
| 2719 |
+
"train_runtime": 9902.9795,
|
| 2720 |
+
"train_samples_per_second": 49.694,
|
| 2721 |
+
"train_steps_per_second": 0.194
|
| 2722 |
+
}
|
| 2723 |
+
],
|
| 2724 |
+
"logging_steps": 5,
|
| 2725 |
+
"max_steps": 1922,
|
| 2726 |
+
"num_input_tokens_seen": 0,
|
| 2727 |
+
"num_train_epochs": 1,
|
| 2728 |
+
"save_steps": 500,
|
| 2729 |
+
"stateful_callbacks": {
|
| 2730 |
+
"TrainerControl": {
|
| 2731 |
+
"args": {
|
| 2732 |
+
"should_epoch_stop": false,
|
| 2733 |
+
"should_evaluate": false,
|
| 2734 |
+
"should_log": false,
|
| 2735 |
+
"should_save": true,
|
| 2736 |
+
"should_training_stop": true
|
| 2737 |
+
},
|
| 2738 |
+
"attributes": {}
|
| 2739 |
+
}
|
| 2740 |
+
},
|
| 2741 |
+
"total_flos": 1058556698689536.0,
|
| 2742 |
+
"train_batch_size": 8,
|
| 2743 |
+
"trial_name": null,
|
| 2744 |
+
"trial_params": null
|
| 2745 |
+
}
|