Upload folder using huggingface_hub
Browse files- README.md +58 -3
- chat_template.jinja +15 -0
- checkpoint-1000/chat_template.jinja +15 -0
- checkpoint-1000/config.json +29 -0
- checkpoint-1000/generation_config.json +7 -0
- checkpoint-1000/model.safetensors +3 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/special_tokens_map.json +30 -0
- checkpoint-1000/tokenizer.json +0 -0
- checkpoint-1000/tokenizer.model +3 -0
- checkpoint-1000/tokenizer_config.json +43 -0
- checkpoint-1000/trainer_state.json +934 -0
- checkpoint-1000/training_args.bin +3 -0
- config.json +29 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +43 -0
- trainer_state.json +934 -0
- training_args.bin +3 -0
README.md
CHANGED
|
@@ -1,3 +1,58 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: TinyLlama-1.1B-Chat-v1.0-sft
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- sft
|
| 8 |
+
- trl
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for TinyLlama-1.1B-Chat-v1.0-sft
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0).
|
| 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="None", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
This model was trained with SFT.
|
| 34 |
+
|
| 35 |
+
### Framework versions
|
| 36 |
+
|
| 37 |
+
- TRL: 0.19.0
|
| 38 |
+
- Transformers: 4.53.0
|
| 39 |
+
- Pytorch: 2.7.1+cu118
|
| 40 |
+
- Datasets: 3.6.0
|
| 41 |
+
- Tokenizers: 0.21.2
|
| 42 |
+
|
| 43 |
+
## Citations
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
Cite TRL as:
|
| 48 |
+
|
| 49 |
+
```bibtex
|
| 50 |
+
@misc{vonwerra2022trl,
|
| 51 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 52 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
|
| 53 |
+
year = 2020,
|
| 54 |
+
journal = {GitHub repository},
|
| 55 |
+
publisher = {GitHub},
|
| 56 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 57 |
+
}
|
| 58 |
+
```
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}
|
| 2 |
+
{% if message['role'] == 'user' %}
|
| 3 |
+
{{ '<|user|>
|
| 4 |
+
' + message['content'] + eos_token }}
|
| 5 |
+
{% elif message['role'] == 'system' %}
|
| 6 |
+
{{ '<|system|>
|
| 7 |
+
' + message['content'] + eos_token }}
|
| 8 |
+
{% elif message['role'] == 'assistant' %}
|
| 9 |
+
{{ '<|assistant|>
|
| 10 |
+
' + message['content'] + eos_token }}
|
| 11 |
+
{% endif %}
|
| 12 |
+
{% if loop.last and add_generation_prompt %}
|
| 13 |
+
{{ '<|assistant|>' }}
|
| 14 |
+
{% endif %}
|
| 15 |
+
{% endfor %}
|
checkpoint-1000/chat_template.jinja
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}
|
| 2 |
+
{% if message['role'] == 'user' %}
|
| 3 |
+
{{ '<|user|>
|
| 4 |
+
' + message['content'] + eos_token }}
|
| 5 |
+
{% elif message['role'] == 'system' %}
|
| 6 |
+
{{ '<|system|>
|
| 7 |
+
' + message['content'] + eos_token }}
|
| 8 |
+
{% elif message['role'] == 'assistant' %}
|
| 9 |
+
{{ '<|assistant|>
|
| 10 |
+
' + message['content'] + eos_token }}
|
| 11 |
+
{% endif %}
|
| 12 |
+
{% if loop.last and add_generation_prompt %}
|
| 13 |
+
{{ '<|assistant|>' }}
|
| 14 |
+
{% endif %}
|
| 15 |
+
{% endfor %}
|
checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"head_dim": 64,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 2048,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 5632,
|
| 14 |
+
"max_position_embeddings": 2048,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
+
"model_type": "llama",
|
| 17 |
+
"num_attention_heads": 32,
|
| 18 |
+
"num_hidden_layers": 22,
|
| 19 |
+
"num_key_value_heads": 4,
|
| 20 |
+
"pretraining_tp": 1,
|
| 21 |
+
"rms_norm_eps": 1e-05,
|
| 22 |
+
"rope_scaling": null,
|
| 23 |
+
"rope_theta": 10000.0,
|
| 24 |
+
"tie_word_embeddings": false,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.53.0",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"vocab_size": 32001
|
| 29 |
+
}
|
checkpoint-1000/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.53.0"
|
| 7 |
+
}
|
checkpoint-1000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ce140b02e2c527fa587492251e6f0c3e30b8fb48c0a0ecaf5a32fec46a01bb7
|
| 3 |
+
size 4400232920
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7087ae7747d0e66edbbfb8de9ae584bec24f5cfce7205ab49f4ff0fccc92309
|
| 3 |
+
size 8800594203
|
checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b5a67da83d9c0e696e53b6ca872b7aedf0300a6f1f06e4a2fda8e6850bba188
|
| 3 |
+
size 14645
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4cf3f593de075f8621f954f65e72b4201caf184063029598ab21043538178a42
|
| 3 |
+
size 1465
|
checkpoint-1000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
checkpoint-1000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1000/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
checkpoint-1000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "</s>",
|
| 34 |
+
"extra_special_tokens": {},
|
| 35 |
+
"legacy": false,
|
| 36 |
+
"model_max_length": 2048,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"padding_side": "right",
|
| 39 |
+
"sp_model_kwargs": {},
|
| 40 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 41 |
+
"unk_token": "<unk>",
|
| 42 |
+
"use_default_system_prompt": false
|
| 43 |
+
}
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,934 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.1952,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.032,
|
| 14 |
+
"grad_norm": 4.924402713775635,
|
| 15 |
+
"learning_rate": 5.69620253164557e-07,
|
| 16 |
+
"loss": 1.003,
|
| 17 |
+
"mean_token_accuracy": 0.7565429553389549,
|
| 18 |
+
"num_tokens": 108351.0,
|
| 19 |
+
"step": 10
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.064,
|
| 23 |
+
"grad_norm": 5.024679183959961,
|
| 24 |
+
"learning_rate": 1.2025316455696204e-06,
|
| 25 |
+
"loss": 0.9631,
|
| 26 |
+
"mean_token_accuracy": 0.7678055994212627,
|
| 27 |
+
"num_tokens": 220746.0,
|
| 28 |
+
"step": 20
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.096,
|
| 32 |
+
"grad_norm": 3.090874671936035,
|
| 33 |
+
"learning_rate": 1.8354430379746838e-06,
|
| 34 |
+
"loss": 0.884,
|
| 35 |
+
"mean_token_accuracy": 0.779016162455082,
|
| 36 |
+
"num_tokens": 333736.0,
|
| 37 |
+
"step": 30
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.128,
|
| 41 |
+
"grad_norm": 2.65234637260437,
|
| 42 |
+
"learning_rate": 2.4683544303797473e-06,
|
| 43 |
+
"loss": 0.7815,
|
| 44 |
+
"mean_token_accuracy": 0.8002516604959965,
|
| 45 |
+
"num_tokens": 449637.0,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.16,
|
| 50 |
+
"grad_norm": 2.4638168811798096,
|
| 51 |
+
"learning_rate": 3.10126582278481e-06,
|
| 52 |
+
"loss": 0.75,
|
| 53 |
+
"mean_token_accuracy": 0.8030893772840499,
|
| 54 |
+
"num_tokens": 562989.0,
|
| 55 |
+
"step": 50
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.192,
|
| 59 |
+
"grad_norm": 2.374603509902954,
|
| 60 |
+
"learning_rate": 3.7341772151898737e-06,
|
| 61 |
+
"loss": 0.7152,
|
| 62 |
+
"mean_token_accuracy": 0.8119513258337975,
|
| 63 |
+
"num_tokens": 673839.0,
|
| 64 |
+
"step": 60
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.224,
|
| 68 |
+
"grad_norm": 2.330502510070801,
|
| 69 |
+
"learning_rate": 4.367088607594937e-06,
|
| 70 |
+
"loss": 0.6941,
|
| 71 |
+
"mean_token_accuracy": 0.815937215834856,
|
| 72 |
+
"num_tokens": 781061.0,
|
| 73 |
+
"step": 70
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.256,
|
| 77 |
+
"grad_norm": 2.267749309539795,
|
| 78 |
+
"learning_rate": 5e-06,
|
| 79 |
+
"loss": 0.696,
|
| 80 |
+
"mean_token_accuracy": 0.8140616729855538,
|
| 81 |
+
"num_tokens": 891312.0,
|
| 82 |
+
"step": 80
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.288,
|
| 86 |
+
"grad_norm": 2.236624240875244,
|
| 87 |
+
"learning_rate": 4.9994413292115925e-06,
|
| 88 |
+
"loss": 0.6568,
|
| 89 |
+
"mean_token_accuracy": 0.8207014128565788,
|
| 90 |
+
"num_tokens": 1005632.0,
|
| 91 |
+
"step": 90
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.32,
|
| 95 |
+
"grad_norm": 2.1301517486572266,
|
| 96 |
+
"learning_rate": 4.997765566536809e-06,
|
| 97 |
+
"loss": 0.6286,
|
| 98 |
+
"mean_token_accuracy": 0.8304322719573974,
|
| 99 |
+
"num_tokens": 1119621.0,
|
| 100 |
+
"step": 100
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.352,
|
| 104 |
+
"grad_norm": 2.2088050842285156,
|
| 105 |
+
"learning_rate": 4.9949734609353746e-06,
|
| 106 |
+
"loss": 0.6173,
|
| 107 |
+
"mean_token_accuracy": 0.8298865057528019,
|
| 108 |
+
"num_tokens": 1228735.0,
|
| 109 |
+
"step": 110
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.384,
|
| 113 |
+
"grad_norm": 2.109389305114746,
|
| 114 |
+
"learning_rate": 4.991066260301558e-06,
|
| 115 |
+
"loss": 0.6392,
|
| 116 |
+
"mean_token_accuracy": 0.8268058508634567,
|
| 117 |
+
"num_tokens": 1339064.0,
|
| 118 |
+
"step": 120
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.416,
|
| 122 |
+
"grad_norm": 1.9944440126419067,
|
| 123 |
+
"learning_rate": 4.986045710906447e-06,
|
| 124 |
+
"loss": 0.5951,
|
| 125 |
+
"mean_token_accuracy": 0.8359164826571941,
|
| 126 |
+
"num_tokens": 1452954.0,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.448,
|
| 131 |
+
"grad_norm": 2.0711472034454346,
|
| 132 |
+
"learning_rate": 4.979914056617472e-06,
|
| 133 |
+
"loss": 0.6136,
|
| 134 |
+
"mean_token_accuracy": 0.8322145372629166,
|
| 135 |
+
"num_tokens": 1564718.0,
|
| 136 |
+
"step": 140
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.48,
|
| 140 |
+
"grad_norm": 2.3020684719085693,
|
| 141 |
+
"learning_rate": 4.972674037895542e-06,
|
| 142 |
+
"loss": 0.593,
|
| 143 |
+
"mean_token_accuracy": 0.8376348964869976,
|
| 144 |
+
"num_tokens": 1673953.0,
|
| 145 |
+
"step": 150
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.512,
|
| 149 |
+
"grad_norm": 2.002972364425659,
|
| 150 |
+
"learning_rate": 4.964328890570231e-06,
|
| 151 |
+
"loss": 0.6062,
|
| 152 |
+
"mean_token_accuracy": 0.8321727059781552,
|
| 153 |
+
"num_tokens": 1780894.0,
|
| 154 |
+
"step": 160
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.544,
|
| 158 |
+
"grad_norm": 2.345299482345581,
|
| 159 |
+
"learning_rate": 4.954882344393566e-06,
|
| 160 |
+
"loss": 0.6018,
|
| 161 |
+
"mean_token_accuracy": 0.8306759320199489,
|
| 162 |
+
"num_tokens": 1895107.0,
|
| 163 |
+
"step": 170
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.576,
|
| 167 |
+
"grad_norm": 2.180422306060791,
|
| 168 |
+
"learning_rate": 4.94433862137307e-06,
|
| 169 |
+
"loss": 0.5945,
|
| 170 |
+
"mean_token_accuracy": 0.834496195614338,
|
| 171 |
+
"num_tokens": 2004705.0,
|
| 172 |
+
"step": 180
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.608,
|
| 176 |
+
"grad_norm": 2.1533453464508057,
|
| 177 |
+
"learning_rate": 4.9327024338847836e-06,
|
| 178 |
+
"loss": 0.5989,
|
| 179 |
+
"mean_token_accuracy": 0.8349006526172161,
|
| 180 |
+
"num_tokens": 2117436.0,
|
| 181 |
+
"step": 190
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.64,
|
| 185 |
+
"grad_norm": 2.128485918045044,
|
| 186 |
+
"learning_rate": 4.919978982567138e-06,
|
| 187 |
+
"loss": 0.5582,
|
| 188 |
+
"mean_token_accuracy": 0.8449515514075756,
|
| 189 |
+
"num_tokens": 2229604.0,
|
| 190 |
+
"step": 200
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.672,
|
| 194 |
+
"grad_norm": 2.323518753051758,
|
| 195 |
+
"learning_rate": 4.906173953996596e-06,
|
| 196 |
+
"loss": 0.5596,
|
| 197 |
+
"mean_token_accuracy": 0.8436533316969872,
|
| 198 |
+
"num_tokens": 2343563.0,
|
| 199 |
+
"step": 210
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.704,
|
| 203 |
+
"grad_norm": 2.2156243324279785,
|
| 204 |
+
"learning_rate": 4.891293518146112e-06,
|
| 205 |
+
"loss": 0.5855,
|
| 206 |
+
"mean_token_accuracy": 0.8377246461808682,
|
| 207 |
+
"num_tokens": 2453154.0,
|
| 208 |
+
"step": 220
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.736,
|
| 212 |
+
"grad_norm": 2.2261300086975098,
|
| 213 |
+
"learning_rate": 4.875344325627551e-06,
|
| 214 |
+
"loss": 0.5716,
|
| 215 |
+
"mean_token_accuracy": 0.8415978990495205,
|
| 216 |
+
"num_tokens": 2566369.0,
|
| 217 |
+
"step": 230
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.768,
|
| 221 |
+
"grad_norm": 2.0974161624908447,
|
| 222 |
+
"learning_rate": 4.85833350471928e-06,
|
| 223 |
+
"loss": 0.5628,
|
| 224 |
+
"mean_token_accuracy": 0.8448635570704937,
|
| 225 |
+
"num_tokens": 2674001.0,
|
| 226 |
+
"step": 240
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8,
|
| 230 |
+
"grad_norm": 2.181368112564087,
|
| 231 |
+
"learning_rate": 4.840268658180281e-06,
|
| 232 |
+
"loss": 0.5575,
|
| 233 |
+
"mean_token_accuracy": 0.8437174156308174,
|
| 234 |
+
"num_tokens": 2785131.0,
|
| 235 |
+
"step": 250
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.832,
|
| 239 |
+
"grad_norm": 1.9916293621063232,
|
| 240 |
+
"learning_rate": 4.821157859852201e-06,
|
| 241 |
+
"loss": 0.5496,
|
| 242 |
+
"mean_token_accuracy": 0.8463803254067898,
|
| 243 |
+
"num_tokens": 2898708.0,
|
| 244 |
+
"step": 260
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.864,
|
| 248 |
+
"grad_norm": 1.9616676568984985,
|
| 249 |
+
"learning_rate": 4.801009651050856e-06,
|
| 250 |
+
"loss": 0.5599,
|
| 251 |
+
"mean_token_accuracy": 0.8417543187737465,
|
| 252 |
+
"num_tokens": 3008577.0,
|
| 253 |
+
"step": 270
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.896,
|
| 257 |
+
"grad_norm": 2.2142021656036377,
|
| 258 |
+
"learning_rate": 4.779833036748801e-06,
|
| 259 |
+
"loss": 0.5268,
|
| 260 |
+
"mean_token_accuracy": 0.8514743112027645,
|
| 261 |
+
"num_tokens": 3120102.0,
|
| 262 |
+
"step": 280
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.928,
|
| 266 |
+
"grad_norm": 1.8409771919250488,
|
| 267 |
+
"learning_rate": 4.757637481550683e-06,
|
| 268 |
+
"loss": 0.5443,
|
| 269 |
+
"mean_token_accuracy": 0.8460176154971123,
|
| 270 |
+
"num_tokens": 3234456.0,
|
| 271 |
+
"step": 290
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.96,
|
| 275 |
+
"grad_norm": 1.9530022144317627,
|
| 276 |
+
"learning_rate": 4.73443290546316e-06,
|
| 277 |
+
"loss": 0.5487,
|
| 278 |
+
"mean_token_accuracy": 0.8462878614664078,
|
| 279 |
+
"num_tokens": 3348821.0,
|
| 280 |
+
"step": 300
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.992,
|
| 284 |
+
"grad_norm": 1.9719116687774658,
|
| 285 |
+
"learning_rate": 4.7102296794612865e-06,
|
| 286 |
+
"loss": 0.5229,
|
| 287 |
+
"mean_token_accuracy": 0.8531442761421204,
|
| 288 |
+
"num_tokens": 3455469.0,
|
| 289 |
+
"step": 310
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0224,
|
| 293 |
+
"grad_norm": 1.9529277086257935,
|
| 294 |
+
"learning_rate": 4.6850386208533424e-06,
|
| 295 |
+
"loss": 0.4849,
|
| 296 |
+
"mean_token_accuracy": 0.8625573249239671,
|
| 297 |
+
"num_tokens": 3561864.0,
|
| 298 |
+
"step": 320
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.0544,
|
| 302 |
+
"grad_norm": 2.105419635772705,
|
| 303 |
+
"learning_rate": 4.658870988446189e-06,
|
| 304 |
+
"loss": 0.4819,
|
| 305 |
+
"mean_token_accuracy": 0.864888047426939,
|
| 306 |
+
"num_tokens": 3672405.0,
|
| 307 |
+
"step": 330
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.0864,
|
| 311 |
+
"grad_norm": 2.035449504852295,
|
| 312 |
+
"learning_rate": 4.631738477513286e-06,
|
| 313 |
+
"loss": 0.4735,
|
| 314 |
+
"mean_token_accuracy": 0.8661263197660446,
|
| 315 |
+
"num_tokens": 3780404.0,
|
| 316 |
+
"step": 340
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.1184,
|
| 320 |
+
"grad_norm": 2.1748783588409424,
|
| 321 |
+
"learning_rate": 4.603653214567654e-06,
|
| 322 |
+
"loss": 0.4638,
|
| 323 |
+
"mean_token_accuracy": 0.8661996729671955,
|
| 324 |
+
"num_tokens": 3894359.0,
|
| 325 |
+
"step": 350
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.1504,
|
| 329 |
+
"grad_norm": 2.0794739723205566,
|
| 330 |
+
"learning_rate": 4.574627751942087e-06,
|
| 331 |
+
"loss": 0.4683,
|
| 332 |
+
"mean_token_accuracy": 0.8664901047945023,
|
| 333 |
+
"num_tokens": 4007235.0,
|
| 334 |
+
"step": 360
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.1824,
|
| 338 |
+
"grad_norm": 2.2591140270233154,
|
| 339 |
+
"learning_rate": 4.544675062179056e-06,
|
| 340 |
+
"loss": 0.4616,
|
| 341 |
+
"mean_token_accuracy": 0.8676725938916207,
|
| 342 |
+
"num_tokens": 4116127.0,
|
| 343 |
+
"step": 370
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.2144,
|
| 347 |
+
"grad_norm": 2.111358165740967,
|
| 348 |
+
"learning_rate": 4.513808532232804e-06,
|
| 349 |
+
"loss": 0.4742,
|
| 350 |
+
"mean_token_accuracy": 0.8660650610923767,
|
| 351 |
+
"num_tokens": 4228679.0,
|
| 352 |
+
"step": 380
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.2464,
|
| 356 |
+
"grad_norm": 2.115079402923584,
|
| 357 |
+
"learning_rate": 4.482041957486229e-06,
|
| 358 |
+
"loss": 0.4679,
|
| 359 |
+
"mean_token_accuracy": 0.8653517335653305,
|
| 360 |
+
"num_tokens": 4341814.0,
|
| 361 |
+
"step": 390
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.2784,
|
| 365 |
+
"grad_norm": 2.15097713470459,
|
| 366 |
+
"learning_rate": 4.4493895355852176e-06,
|
| 367 |
+
"loss": 0.448,
|
| 368 |
+
"mean_token_accuracy": 0.8716798067092896,
|
| 369 |
+
"num_tokens": 4451355.0,
|
| 370 |
+
"step": 400
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.3104,
|
| 374 |
+
"grad_norm": 2.124791383743286,
|
| 375 |
+
"learning_rate": 4.415865860093199e-06,
|
| 376 |
+
"loss": 0.4576,
|
| 377 |
+
"mean_token_accuracy": 0.8684222124516964,
|
| 378 |
+
"num_tokens": 4566924.0,
|
| 379 |
+
"step": 410
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 1.3424,
|
| 383 |
+
"grad_norm": 2.033034563064575,
|
| 384 |
+
"learning_rate": 4.381485913968747e-06,
|
| 385 |
+
"loss": 0.4558,
|
| 386 |
+
"mean_token_accuracy": 0.8672670938074589,
|
| 387 |
+
"num_tokens": 4677684.0,
|
| 388 |
+
"step": 420
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.3744,
|
| 392 |
+
"grad_norm": 2.179928779602051,
|
| 393 |
+
"learning_rate": 4.346265062869147e-06,
|
| 394 |
+
"loss": 0.468,
|
| 395 |
+
"mean_token_accuracy": 0.8666120313107968,
|
| 396 |
+
"num_tokens": 4787468.0,
|
| 397 |
+
"step": 430
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.4064,
|
| 401 |
+
"grad_norm": 2.1226675510406494,
|
| 402 |
+
"learning_rate": 4.310219048282921e-06,
|
| 403 |
+
"loss": 0.4514,
|
| 404 |
+
"mean_token_accuracy": 0.8693858951330184,
|
| 405 |
+
"num_tokens": 4899692.0,
|
| 406 |
+
"step": 440
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.4384000000000001,
|
| 410 |
+
"grad_norm": 2.226494550704956,
|
| 411 |
+
"learning_rate": 4.273363980494379e-06,
|
| 412 |
+
"loss": 0.4761,
|
| 413 |
+
"mean_token_accuracy": 0.8673883900046349,
|
| 414 |
+
"num_tokens": 5008859.0,
|
| 415 |
+
"step": 450
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.4704,
|
| 419 |
+
"grad_norm": 2.039381980895996,
|
| 420 |
+
"learning_rate": 4.235716331383343e-06,
|
| 421 |
+
"loss": 0.4652,
|
| 422 |
+
"mean_token_accuracy": 0.8683492943644524,
|
| 423 |
+
"num_tokens": 5117719.0,
|
| 424 |
+
"step": 460
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.5024,
|
| 428 |
+
"grad_norm": 2.1596784591674805,
|
| 429 |
+
"learning_rate": 4.197292927063263e-06,
|
| 430 |
+
"loss": 0.4701,
|
| 431 |
+
"mean_token_accuracy": 0.863686279207468,
|
| 432 |
+
"num_tokens": 5227041.0,
|
| 433 |
+
"step": 470
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.5344,
|
| 437 |
+
"grad_norm": 2.153256893157959,
|
| 438 |
+
"learning_rate": 4.158110940361007e-06,
|
| 439 |
+
"loss": 0.4559,
|
| 440 |
+
"mean_token_accuracy": 0.8705666847527027,
|
| 441 |
+
"num_tokens": 5334777.0,
|
| 442 |
+
"step": 480
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 1.5664,
|
| 446 |
+
"grad_norm": 2.5348920822143555,
|
| 447 |
+
"learning_rate": 4.118187883141694e-06,
|
| 448 |
+
"loss": 0.461,
|
| 449 |
+
"mean_token_accuracy": 0.8692857407033443,
|
| 450 |
+
"num_tokens": 5441237.0,
|
| 451 |
+
"step": 490
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.5984,
|
| 455 |
+
"grad_norm": 2.2534334659576416,
|
| 456 |
+
"learning_rate": 4.077541598482009e-06,
|
| 457 |
+
"loss": 0.4748,
|
| 458 |
+
"mean_token_accuracy": 0.8672291107475758,
|
| 459 |
+
"num_tokens": 5553620.0,
|
| 460 |
+
"step": 500
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.6303999999999998,
|
| 464 |
+
"grad_norm": 2.1640467643737793,
|
| 465 |
+
"learning_rate": 4.036190252695467e-06,
|
| 466 |
+
"loss": 0.4874,
|
| 467 |
+
"mean_token_accuracy": 0.8607077576220036,
|
| 468 |
+
"num_tokens": 5663373.0,
|
| 469 |
+
"step": 510
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.6623999999999999,
|
| 473 |
+
"grad_norm": 2.1866941452026367,
|
| 474 |
+
"learning_rate": 3.994152327213232e-06,
|
| 475 |
+
"loss": 0.4545,
|
| 476 |
+
"mean_token_accuracy": 0.8709075525403023,
|
| 477 |
+
"num_tokens": 5776290.0,
|
| 478 |
+
"step": 520
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.6944,
|
| 482 |
+
"grad_norm": 2.4210782051086426,
|
| 483 |
+
"learning_rate": 3.95144661032408e-06,
|
| 484 |
+
"loss": 0.4316,
|
| 485 |
+
"mean_token_accuracy": 0.8738773807883262,
|
| 486 |
+
"num_tokens": 5882282.0,
|
| 487 |
+
"step": 530
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.7264,
|
| 491 |
+
"grad_norm": 2.2260217666625977,
|
| 492 |
+
"learning_rate": 3.9080921887772314e-06,
|
| 493 |
+
"loss": 0.4497,
|
| 494 |
+
"mean_token_accuracy": 0.8696616031229496,
|
| 495 |
+
"num_tokens": 5999645.0,
|
| 496 |
+
"step": 540
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.7584,
|
| 500 |
+
"grad_norm": 2.1372995376586914,
|
| 501 |
+
"learning_rate": 3.864108439251779e-06,
|
| 502 |
+
"loss": 0.4466,
|
| 503 |
+
"mean_token_accuracy": 0.8693803034722805,
|
| 504 |
+
"num_tokens": 6108677.0,
|
| 505 |
+
"step": 550
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.7904,
|
| 509 |
+
"grad_norm": 2.0875401496887207,
|
| 510 |
+
"learning_rate": 3.8195150196965414e-06,
|
| 511 |
+
"loss": 0.4618,
|
| 512 |
+
"mean_token_accuracy": 0.866703423857689,
|
| 513 |
+
"num_tokens": 6222393.0,
|
| 514 |
+
"step": 560
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.8224,
|
| 518 |
+
"grad_norm": 2.06887149810791,
|
| 519 |
+
"learning_rate": 3.7743318605442093e-06,
|
| 520 |
+
"loss": 0.4593,
|
| 521 |
+
"mean_token_accuracy": 0.8678097896277904,
|
| 522 |
+
"num_tokens": 6337550.0,
|
| 523 |
+
"step": 570
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 1.8544,
|
| 527 |
+
"grad_norm": 1.9830808639526367,
|
| 528 |
+
"learning_rate": 3.728579155803697e-06,
|
| 529 |
+
"loss": 0.4577,
|
| 530 |
+
"mean_token_accuracy": 0.8675460621714592,
|
| 531 |
+
"num_tokens": 6449551.0,
|
| 532 |
+
"step": 580
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.8864,
|
| 536 |
+
"grad_norm": 2.165144443511963,
|
| 537 |
+
"learning_rate": 3.6822773540347107e-06,
|
| 538 |
+
"loss": 0.439,
|
| 539 |
+
"mean_token_accuracy": 0.8739401362836361,
|
| 540 |
+
"num_tokens": 6562113.0,
|
| 541 |
+
"step": 590
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.9184,
|
| 545 |
+
"grad_norm": 2.1738622188568115,
|
| 546 |
+
"learning_rate": 3.635447149208528e-06,
|
| 547 |
+
"loss": 0.4412,
|
| 548 |
+
"mean_token_accuracy": 0.8708230577409267,
|
| 549 |
+
"num_tokens": 6674860.0,
|
| 550 |
+
"step": 600
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.9504000000000001,
|
| 554 |
+
"grad_norm": 2.4379212856292725,
|
| 555 |
+
"learning_rate": 3.5881094714591102e-06,
|
| 556 |
+
"loss": 0.4522,
|
| 557 |
+
"mean_token_accuracy": 0.8691440485417843,
|
| 558 |
+
"num_tokens": 6789358.0,
|
| 559 |
+
"step": 610
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.9824000000000002,
|
| 563 |
+
"grad_norm": 2.1988134384155273,
|
| 564 |
+
"learning_rate": 3.540285477728658e-06,
|
| 565 |
+
"loss": 0.4351,
|
| 566 |
+
"mean_token_accuracy": 0.8745189763605594,
|
| 567 |
+
"num_tokens": 6899897.0,
|
| 568 |
+
"step": 620
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.0128,
|
| 572 |
+
"grad_norm": 2.226576089859009,
|
| 573 |
+
"learning_rate": 3.491996542311795e-06,
|
| 574 |
+
"loss": 0.4196,
|
| 575 |
+
"mean_token_accuracy": 0.8793611534331974,
|
| 576 |
+
"num_tokens": 7005635.0,
|
| 577 |
+
"step": 630
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 2.0448,
|
| 581 |
+
"grad_norm": 2.284040927886963,
|
| 582 |
+
"learning_rate": 3.44326424730262e-06,
|
| 583 |
+
"loss": 0.3729,
|
| 584 |
+
"mean_token_accuracy": 0.8900357685983181,
|
| 585 |
+
"num_tokens": 7118882.0,
|
| 586 |
+
"step": 640
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.0768,
|
| 590 |
+
"grad_norm": 2.3776021003723145,
|
| 591 |
+
"learning_rate": 3.394110372948871e-06,
|
| 592 |
+
"loss": 0.3784,
|
| 593 |
+
"mean_token_accuracy": 0.8912972524762154,
|
| 594 |
+
"num_tokens": 7229223.0,
|
| 595 |
+
"step": 650
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 2.1088,
|
| 599 |
+
"grad_norm": 2.3752377033233643,
|
| 600 |
+
"learning_rate": 3.344556887917539e-06,
|
| 601 |
+
"loss": 0.3754,
|
| 602 |
+
"mean_token_accuracy": 0.8933634988963604,
|
| 603 |
+
"num_tokens": 7344352.0,
|
| 604 |
+
"step": 660
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.1408,
|
| 608 |
+
"grad_norm": 2.2204787731170654,
|
| 609 |
+
"learning_rate": 3.2946259394762645e-06,
|
| 610 |
+
"loss": 0.3603,
|
| 611 |
+
"mean_token_accuracy": 0.8949612870812416,
|
| 612 |
+
"num_tokens": 7452978.0,
|
| 613 |
+
"step": 670
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.1728,
|
| 617 |
+
"grad_norm": 2.3392176628112793,
|
| 618 |
+
"learning_rate": 3.244339843594913e-06,
|
| 619 |
+
"loss": 0.3633,
|
| 620 |
+
"mean_token_accuracy": 0.8934120312333107,
|
| 621 |
+
"num_tokens": 7562987.0,
|
| 622 |
+
"step": 680
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.2048,
|
| 626 |
+
"grad_norm": 2.548868417739868,
|
| 627 |
+
"learning_rate": 3.19372107497175e-06,
|
| 628 |
+
"loss": 0.3688,
|
| 629 |
+
"mean_token_accuracy": 0.8925295993685722,
|
| 630 |
+
"num_tokens": 7672018.0,
|
| 631 |
+
"step": 690
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 2.2368,
|
| 635 |
+
"grad_norm": 2.524747371673584,
|
| 636 |
+
"learning_rate": 3.1427922569886756e-06,
|
| 637 |
+
"loss": 0.3535,
|
| 638 |
+
"mean_token_accuracy": 0.8959250211715698,
|
| 639 |
+
"num_tokens": 7778694.0,
|
| 640 |
+
"step": 700
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.2688,
|
| 644 |
+
"grad_norm": 2.2159647941589355,
|
| 645 |
+
"learning_rate": 3.0915761516000053e-06,
|
| 646 |
+
"loss": 0.3645,
|
| 647 |
+
"mean_token_accuracy": 0.8916135348379612,
|
| 648 |
+
"num_tokens": 7889175.0,
|
| 649 |
+
"step": 710
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 2.3008,
|
| 653 |
+
"grad_norm": 2.6435368061065674,
|
| 654 |
+
"learning_rate": 3.0400956491593215e-06,
|
| 655 |
+
"loss": 0.3742,
|
| 656 |
+
"mean_token_accuracy": 0.8912425316870213,
|
| 657 |
+
"num_tokens": 7997644.0,
|
| 658 |
+
"step": 720
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 2.3327999999999998,
|
| 662 |
+
"grad_norm": 2.5125110149383545,
|
| 663 |
+
"learning_rate": 2.9883737581889315e-06,
|
| 664 |
+
"loss": 0.3767,
|
| 665 |
+
"mean_token_accuracy": 0.8903645075857639,
|
| 666 |
+
"num_tokens": 8109860.0,
|
| 667 |
+
"step": 730
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.3648,
|
| 671 |
+
"grad_norm": 2.482232093811035,
|
| 672 |
+
"learning_rate": 2.9364335950965227e-06,
|
| 673 |
+
"loss": 0.3531,
|
| 674 |
+
"mean_token_accuracy": 0.8950154200196266,
|
| 675 |
+
"num_tokens": 8219541.0,
|
| 676 |
+
"step": 740
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.3968,
|
| 680 |
+
"grad_norm": 2.360308885574341,
|
| 681 |
+
"learning_rate": 2.8842983738435855e-06,
|
| 682 |
+
"loss": 0.3602,
|
| 683 |
+
"mean_token_accuracy": 0.892794082313776,
|
| 684 |
+
"num_tokens": 8329765.0,
|
| 685 |
+
"step": 750
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 2.4288,
|
| 689 |
+
"grad_norm": 2.5826432704925537,
|
| 690 |
+
"learning_rate": 2.831991395570249e-06,
|
| 691 |
+
"loss": 0.364,
|
| 692 |
+
"mean_token_accuracy": 0.8930977649986744,
|
| 693 |
+
"num_tokens": 8443917.0,
|
| 694 |
+
"step": 760
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 2.4608,
|
| 698 |
+
"grad_norm": 2.588911771774292,
|
| 699 |
+
"learning_rate": 2.779536038181146e-06,
|
| 700 |
+
"loss": 0.3771,
|
| 701 |
+
"mean_token_accuracy": 0.8915269665420056,
|
| 702 |
+
"num_tokens": 8550550.0,
|
| 703 |
+
"step": 770
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.4928,
|
| 707 |
+
"grad_norm": 2.4379560947418213,
|
| 708 |
+
"learning_rate": 2.726955745896972e-06,
|
| 709 |
+
"loss": 0.3726,
|
| 710 |
+
"mean_token_accuracy": 0.8931849867105484,
|
| 711 |
+
"num_tokens": 8659752.0,
|
| 712 |
+
"step": 780
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.5248,
|
| 716 |
+
"grad_norm": 2.285378932952881,
|
| 717 |
+
"learning_rate": 2.6742740187764026e-06,
|
| 718 |
+
"loss": 0.3644,
|
| 719 |
+
"mean_token_accuracy": 0.8937488354742527,
|
| 720 |
+
"num_tokens": 8775392.0,
|
| 721 |
+
"step": 790
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.5568,
|
| 725 |
+
"grad_norm": 2.6339035034179688,
|
| 726 |
+
"learning_rate": 2.621514402213059e-06,
|
| 727 |
+
"loss": 0.3528,
|
| 728 |
+
"mean_token_accuracy": 0.8967724144458771,
|
| 729 |
+
"num_tokens": 8888499.0,
|
| 730 |
+
"step": 800
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.5888,
|
| 734 |
+
"grad_norm": 2.2561960220336914,
|
| 735 |
+
"learning_rate": 2.568700476412204e-06,
|
| 736 |
+
"loss": 0.3846,
|
| 737 |
+
"mean_token_accuracy": 0.8879048563539982,
|
| 738 |
+
"num_tokens": 9001912.0,
|
| 739 |
+
"step": 810
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.6208,
|
| 743 |
+
"grad_norm": 2.3487298488616943,
|
| 744 |
+
"learning_rate": 2.515855845851894e-06,
|
| 745 |
+
"loss": 0.3602,
|
| 746 |
+
"mean_token_accuracy": 0.8923349373042584,
|
| 747 |
+
"num_tokens": 9113468.0,
|
| 748 |
+
"step": 820
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.6528,
|
| 752 |
+
"grad_norm": 2.4302189350128174,
|
| 753 |
+
"learning_rate": 2.46300412873326e-06,
|
| 754 |
+
"loss": 0.3613,
|
| 755 |
+
"mean_token_accuracy": 0.8928801603615284,
|
| 756 |
+
"num_tokens": 9229335.0,
|
| 757 |
+
"step": 830
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 2.6848,
|
| 761 |
+
"grad_norm": 2.48608136177063,
|
| 762 |
+
"learning_rate": 2.4101689464246816e-06,
|
| 763 |
+
"loss": 0.3642,
|
| 764 |
+
"mean_token_accuracy": 0.8935355007648468,
|
| 765 |
+
"num_tokens": 9344595.0,
|
| 766 |
+
"step": 840
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.7168,
|
| 770 |
+
"grad_norm": 2.357041120529175,
|
| 771 |
+
"learning_rate": 2.3573739129045227e-06,
|
| 772 |
+
"loss": 0.3579,
|
| 773 |
+
"mean_token_accuracy": 0.8953744530677795,
|
| 774 |
+
"num_tokens": 9455015.0,
|
| 775 |
+
"step": 850
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 2.7488,
|
| 779 |
+
"grad_norm": 2.649960517883301,
|
| 780 |
+
"learning_rate": 2.3046426242071834e-06,
|
| 781 |
+
"loss": 0.3689,
|
| 782 |
+
"mean_token_accuracy": 0.8912502139806747,
|
| 783 |
+
"num_tokens": 9560806.0,
|
| 784 |
+
"step": 860
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 2.7808,
|
| 788 |
+
"grad_norm": 2.3866405487060547,
|
| 789 |
+
"learning_rate": 2.251998647877169e-06,
|
| 790 |
+
"loss": 0.364,
|
| 791 |
+
"mean_token_accuracy": 0.8922963574528694,
|
| 792 |
+
"num_tokens": 9673310.0,
|
| 793 |
+
"step": 870
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.8128,
|
| 797 |
+
"grad_norm": 2.3306539058685303,
|
| 798 |
+
"learning_rate": 2.1994655124358875e-06,
|
| 799 |
+
"loss": 0.347,
|
| 800 |
+
"mean_token_accuracy": 0.8968791022896767,
|
| 801 |
+
"num_tokens": 9787134.0,
|
| 802 |
+
"step": 880
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.8448,
|
| 806 |
+
"grad_norm": 2.3708155155181885,
|
| 807 |
+
"learning_rate": 2.1470666968658956e-06,
|
| 808 |
+
"loss": 0.3576,
|
| 809 |
+
"mean_token_accuracy": 0.8956275820732117,
|
| 810 |
+
"num_tokens": 9901311.0,
|
| 811 |
+
"step": 890
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 2.8768000000000002,
|
| 815 |
+
"grad_norm": 2.466033935546875,
|
| 816 |
+
"learning_rate": 2.0948256201172772e-06,
|
| 817 |
+
"loss": 0.3495,
|
| 818 |
+
"mean_token_accuracy": 0.8957146309316159,
|
| 819 |
+
"num_tokens": 10017025.0,
|
| 820 |
+
"step": 900
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.9088000000000003,
|
| 824 |
+
"grad_norm": 2.630037784576416,
|
| 825 |
+
"learning_rate": 2.0427656306408594e-06,
|
| 826 |
+
"loss": 0.3643,
|
| 827 |
+
"mean_token_accuracy": 0.892961747199297,
|
| 828 |
+
"num_tokens": 10131105.0,
|
| 829 |
+
"step": 910
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 2.9408,
|
| 833 |
+
"grad_norm": 2.4972527027130127,
|
| 834 |
+
"learning_rate": 1.990909995952936e-06,
|
| 835 |
+
"loss": 0.3542,
|
| 836 |
+
"mean_token_accuracy": 0.8959409989416599,
|
| 837 |
+
"num_tokens": 10240555.0,
|
| 838 |
+
"step": 920
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 2.9728,
|
| 842 |
+
"grad_norm": 2.722083330154419,
|
| 843 |
+
"learning_rate": 1.939281892236157e-06,
|
| 844 |
+
"loss": 0.3531,
|
| 845 |
+
"mean_token_accuracy": 0.8945782586932183,
|
| 846 |
+
"num_tokens": 10347889.0,
|
| 847 |
+
"step": 930
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 3.0032,
|
| 851 |
+
"grad_norm": 2.2683236598968506,
|
| 852 |
+
"learning_rate": 1.8879043939812503e-06,
|
| 853 |
+
"loss": 0.3579,
|
| 854 |
+
"mean_token_accuracy": 0.8970388119157992,
|
| 855 |
+
"num_tokens": 10453648.0,
|
| 856 |
+
"step": 940
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 3.0352,
|
| 860 |
+
"grad_norm": 2.5510222911834717,
|
| 861 |
+
"learning_rate": 1.8368004636741794e-06,
|
| 862 |
+
"loss": 0.3062,
|
| 863 |
+
"mean_token_accuracy": 0.9120431691408157,
|
| 864 |
+
"num_tokens": 10565156.0,
|
| 865 |
+
"step": 950
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 3.0672,
|
| 869 |
+
"grad_norm": 2.730621337890625,
|
| 870 |
+
"learning_rate": 1.7859929415333725e-06,
|
| 871 |
+
"loss": 0.2912,
|
| 872 |
+
"mean_token_accuracy": 0.9148132555186749,
|
| 873 |
+
"num_tokens": 10676677.0,
|
| 874 |
+
"step": 960
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 3.0992,
|
| 878 |
+
"grad_norm": 2.5102102756500244,
|
| 879 |
+
"learning_rate": 1.735504535301592e-06,
|
| 880 |
+
"loss": 0.2882,
|
| 881 |
+
"mean_token_accuracy": 0.9159041397273541,
|
| 882 |
+
"num_tokens": 10787468.0,
|
| 883 |
+
"step": 970
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 3.1312,
|
| 887 |
+
"grad_norm": 2.3301403522491455,
|
| 888 |
+
"learning_rate": 1.6853578100970086e-06,
|
| 889 |
+
"loss": 0.3016,
|
| 890 |
+
"mean_token_accuracy": 0.911161607503891,
|
| 891 |
+
"num_tokens": 10897279.0,
|
| 892 |
+
"step": 980
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 3.1632,
|
| 896 |
+
"grad_norm": 2.4964194297790527,
|
| 897 |
+
"learning_rate": 1.6355751783280283e-06,
|
| 898 |
+
"loss": 0.2927,
|
| 899 |
+
"mean_token_accuracy": 0.9142641559243202,
|
| 900 |
+
"num_tokens": 11008731.0,
|
| 901 |
+
"step": 990
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 3.1952,
|
| 905 |
+
"grad_norm": 2.547372341156006,
|
| 906 |
+
"learning_rate": 1.5861788896763625e-06,
|
| 907 |
+
"loss": 0.2991,
|
| 908 |
+
"mean_token_accuracy": 0.9131025090813637,
|
| 909 |
+
"num_tokens": 11123820.0,
|
| 910 |
+
"step": 1000
|
| 911 |
+
}
|
| 912 |
+
],
|
| 913 |
+
"logging_steps": 10,
|
| 914 |
+
"max_steps": 1565,
|
| 915 |
+
"num_input_tokens_seen": 0,
|
| 916 |
+
"num_train_epochs": 5,
|
| 917 |
+
"save_steps": 500,
|
| 918 |
+
"stateful_callbacks": {
|
| 919 |
+
"TrainerControl": {
|
| 920 |
+
"args": {
|
| 921 |
+
"should_epoch_stop": false,
|
| 922 |
+
"should_evaluate": false,
|
| 923 |
+
"should_log": false,
|
| 924 |
+
"should_save": true,
|
| 925 |
+
"should_training_stop": false
|
| 926 |
+
},
|
| 927 |
+
"attributes": {}
|
| 928 |
+
}
|
| 929 |
+
},
|
| 930 |
+
"total_flos": 1.0242157166877082e+17,
|
| 931 |
+
"train_batch_size": 4,
|
| 932 |
+
"trial_name": null,
|
| 933 |
+
"trial_params": null
|
| 934 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:805391ab265a423fa4bf8e6272db7bca4771272624226f8273e5a50efd4ae604
|
| 3 |
+
size 6161
|
config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"head_dim": 64,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 2048,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 5632,
|
| 14 |
+
"max_position_embeddings": 2048,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
+
"model_type": "llama",
|
| 17 |
+
"num_attention_heads": 32,
|
| 18 |
+
"num_hidden_layers": 22,
|
| 19 |
+
"num_key_value_heads": 4,
|
| 20 |
+
"pretraining_tp": 1,
|
| 21 |
+
"rms_norm_eps": 1e-05,
|
| 22 |
+
"rope_scaling": null,
|
| 23 |
+
"rope_theta": 10000.0,
|
| 24 |
+
"tie_word_embeddings": false,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.53.0",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"vocab_size": 32001
|
| 29 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.53.0"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ce140b02e2c527fa587492251e6f0c3e30b8fb48c0a0ecaf5a32fec46a01bb7
|
| 3 |
+
size 4400232920
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7087ae7747d0e66edbbfb8de9ae584bec24f5cfce7205ab49f4ff0fccc92309
|
| 3 |
+
size 8800594203
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b5a67da83d9c0e696e53b6ca872b7aedf0300a6f1f06e4a2fda8e6850bba188
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4cf3f593de075f8621f954f65e72b4201caf184063029598ab21043538178a42
|
| 3 |
+
size 1465
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "</s>",
|
| 34 |
+
"extra_special_tokens": {},
|
| 35 |
+
"legacy": false,
|
| 36 |
+
"model_max_length": 2048,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"padding_side": "right",
|
| 39 |
+
"sp_model_kwargs": {},
|
| 40 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 41 |
+
"unk_token": "<unk>",
|
| 42 |
+
"use_default_system_prompt": false
|
| 43 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,934 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.1952,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.032,
|
| 14 |
+
"grad_norm": 4.924402713775635,
|
| 15 |
+
"learning_rate": 5.69620253164557e-07,
|
| 16 |
+
"loss": 1.003,
|
| 17 |
+
"mean_token_accuracy": 0.7565429553389549,
|
| 18 |
+
"num_tokens": 108351.0,
|
| 19 |
+
"step": 10
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.064,
|
| 23 |
+
"grad_norm": 5.024679183959961,
|
| 24 |
+
"learning_rate": 1.2025316455696204e-06,
|
| 25 |
+
"loss": 0.9631,
|
| 26 |
+
"mean_token_accuracy": 0.7678055994212627,
|
| 27 |
+
"num_tokens": 220746.0,
|
| 28 |
+
"step": 20
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.096,
|
| 32 |
+
"grad_norm": 3.090874671936035,
|
| 33 |
+
"learning_rate": 1.8354430379746838e-06,
|
| 34 |
+
"loss": 0.884,
|
| 35 |
+
"mean_token_accuracy": 0.779016162455082,
|
| 36 |
+
"num_tokens": 333736.0,
|
| 37 |
+
"step": 30
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.128,
|
| 41 |
+
"grad_norm": 2.65234637260437,
|
| 42 |
+
"learning_rate": 2.4683544303797473e-06,
|
| 43 |
+
"loss": 0.7815,
|
| 44 |
+
"mean_token_accuracy": 0.8002516604959965,
|
| 45 |
+
"num_tokens": 449637.0,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.16,
|
| 50 |
+
"grad_norm": 2.4638168811798096,
|
| 51 |
+
"learning_rate": 3.10126582278481e-06,
|
| 52 |
+
"loss": 0.75,
|
| 53 |
+
"mean_token_accuracy": 0.8030893772840499,
|
| 54 |
+
"num_tokens": 562989.0,
|
| 55 |
+
"step": 50
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.192,
|
| 59 |
+
"grad_norm": 2.374603509902954,
|
| 60 |
+
"learning_rate": 3.7341772151898737e-06,
|
| 61 |
+
"loss": 0.7152,
|
| 62 |
+
"mean_token_accuracy": 0.8119513258337975,
|
| 63 |
+
"num_tokens": 673839.0,
|
| 64 |
+
"step": 60
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.224,
|
| 68 |
+
"grad_norm": 2.330502510070801,
|
| 69 |
+
"learning_rate": 4.367088607594937e-06,
|
| 70 |
+
"loss": 0.6941,
|
| 71 |
+
"mean_token_accuracy": 0.815937215834856,
|
| 72 |
+
"num_tokens": 781061.0,
|
| 73 |
+
"step": 70
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.256,
|
| 77 |
+
"grad_norm": 2.267749309539795,
|
| 78 |
+
"learning_rate": 5e-06,
|
| 79 |
+
"loss": 0.696,
|
| 80 |
+
"mean_token_accuracy": 0.8140616729855538,
|
| 81 |
+
"num_tokens": 891312.0,
|
| 82 |
+
"step": 80
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.288,
|
| 86 |
+
"grad_norm": 2.236624240875244,
|
| 87 |
+
"learning_rate": 4.9994413292115925e-06,
|
| 88 |
+
"loss": 0.6568,
|
| 89 |
+
"mean_token_accuracy": 0.8207014128565788,
|
| 90 |
+
"num_tokens": 1005632.0,
|
| 91 |
+
"step": 90
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.32,
|
| 95 |
+
"grad_norm": 2.1301517486572266,
|
| 96 |
+
"learning_rate": 4.997765566536809e-06,
|
| 97 |
+
"loss": 0.6286,
|
| 98 |
+
"mean_token_accuracy": 0.8304322719573974,
|
| 99 |
+
"num_tokens": 1119621.0,
|
| 100 |
+
"step": 100
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.352,
|
| 104 |
+
"grad_norm": 2.2088050842285156,
|
| 105 |
+
"learning_rate": 4.9949734609353746e-06,
|
| 106 |
+
"loss": 0.6173,
|
| 107 |
+
"mean_token_accuracy": 0.8298865057528019,
|
| 108 |
+
"num_tokens": 1228735.0,
|
| 109 |
+
"step": 110
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.384,
|
| 113 |
+
"grad_norm": 2.109389305114746,
|
| 114 |
+
"learning_rate": 4.991066260301558e-06,
|
| 115 |
+
"loss": 0.6392,
|
| 116 |
+
"mean_token_accuracy": 0.8268058508634567,
|
| 117 |
+
"num_tokens": 1339064.0,
|
| 118 |
+
"step": 120
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.416,
|
| 122 |
+
"grad_norm": 1.9944440126419067,
|
| 123 |
+
"learning_rate": 4.986045710906447e-06,
|
| 124 |
+
"loss": 0.5951,
|
| 125 |
+
"mean_token_accuracy": 0.8359164826571941,
|
| 126 |
+
"num_tokens": 1452954.0,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.448,
|
| 131 |
+
"grad_norm": 2.0711472034454346,
|
| 132 |
+
"learning_rate": 4.979914056617472e-06,
|
| 133 |
+
"loss": 0.6136,
|
| 134 |
+
"mean_token_accuracy": 0.8322145372629166,
|
| 135 |
+
"num_tokens": 1564718.0,
|
| 136 |
+
"step": 140
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.48,
|
| 140 |
+
"grad_norm": 2.3020684719085693,
|
| 141 |
+
"learning_rate": 4.972674037895542e-06,
|
| 142 |
+
"loss": 0.593,
|
| 143 |
+
"mean_token_accuracy": 0.8376348964869976,
|
| 144 |
+
"num_tokens": 1673953.0,
|
| 145 |
+
"step": 150
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.512,
|
| 149 |
+
"grad_norm": 2.002972364425659,
|
| 150 |
+
"learning_rate": 4.964328890570231e-06,
|
| 151 |
+
"loss": 0.6062,
|
| 152 |
+
"mean_token_accuracy": 0.8321727059781552,
|
| 153 |
+
"num_tokens": 1780894.0,
|
| 154 |
+
"step": 160
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.544,
|
| 158 |
+
"grad_norm": 2.345299482345581,
|
| 159 |
+
"learning_rate": 4.954882344393566e-06,
|
| 160 |
+
"loss": 0.6018,
|
| 161 |
+
"mean_token_accuracy": 0.8306759320199489,
|
| 162 |
+
"num_tokens": 1895107.0,
|
| 163 |
+
"step": 170
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.576,
|
| 167 |
+
"grad_norm": 2.180422306060791,
|
| 168 |
+
"learning_rate": 4.94433862137307e-06,
|
| 169 |
+
"loss": 0.5945,
|
| 170 |
+
"mean_token_accuracy": 0.834496195614338,
|
| 171 |
+
"num_tokens": 2004705.0,
|
| 172 |
+
"step": 180
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.608,
|
| 176 |
+
"grad_norm": 2.1533453464508057,
|
| 177 |
+
"learning_rate": 4.9327024338847836e-06,
|
| 178 |
+
"loss": 0.5989,
|
| 179 |
+
"mean_token_accuracy": 0.8349006526172161,
|
| 180 |
+
"num_tokens": 2117436.0,
|
| 181 |
+
"step": 190
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.64,
|
| 185 |
+
"grad_norm": 2.128485918045044,
|
| 186 |
+
"learning_rate": 4.919978982567138e-06,
|
| 187 |
+
"loss": 0.5582,
|
| 188 |
+
"mean_token_accuracy": 0.8449515514075756,
|
| 189 |
+
"num_tokens": 2229604.0,
|
| 190 |
+
"step": 200
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.672,
|
| 194 |
+
"grad_norm": 2.323518753051758,
|
| 195 |
+
"learning_rate": 4.906173953996596e-06,
|
| 196 |
+
"loss": 0.5596,
|
| 197 |
+
"mean_token_accuracy": 0.8436533316969872,
|
| 198 |
+
"num_tokens": 2343563.0,
|
| 199 |
+
"step": 210
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.704,
|
| 203 |
+
"grad_norm": 2.2156243324279785,
|
| 204 |
+
"learning_rate": 4.891293518146112e-06,
|
| 205 |
+
"loss": 0.5855,
|
| 206 |
+
"mean_token_accuracy": 0.8377246461808682,
|
| 207 |
+
"num_tokens": 2453154.0,
|
| 208 |
+
"step": 220
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.736,
|
| 212 |
+
"grad_norm": 2.2261300086975098,
|
| 213 |
+
"learning_rate": 4.875344325627551e-06,
|
| 214 |
+
"loss": 0.5716,
|
| 215 |
+
"mean_token_accuracy": 0.8415978990495205,
|
| 216 |
+
"num_tokens": 2566369.0,
|
| 217 |
+
"step": 230
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.768,
|
| 221 |
+
"grad_norm": 2.0974161624908447,
|
| 222 |
+
"learning_rate": 4.85833350471928e-06,
|
| 223 |
+
"loss": 0.5628,
|
| 224 |
+
"mean_token_accuracy": 0.8448635570704937,
|
| 225 |
+
"num_tokens": 2674001.0,
|
| 226 |
+
"step": 240
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8,
|
| 230 |
+
"grad_norm": 2.181368112564087,
|
| 231 |
+
"learning_rate": 4.840268658180281e-06,
|
| 232 |
+
"loss": 0.5575,
|
| 233 |
+
"mean_token_accuracy": 0.8437174156308174,
|
| 234 |
+
"num_tokens": 2785131.0,
|
| 235 |
+
"step": 250
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.832,
|
| 239 |
+
"grad_norm": 1.9916293621063232,
|
| 240 |
+
"learning_rate": 4.821157859852201e-06,
|
| 241 |
+
"loss": 0.5496,
|
| 242 |
+
"mean_token_accuracy": 0.8463803254067898,
|
| 243 |
+
"num_tokens": 2898708.0,
|
| 244 |
+
"step": 260
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.864,
|
| 248 |
+
"grad_norm": 1.9616676568984985,
|
| 249 |
+
"learning_rate": 4.801009651050856e-06,
|
| 250 |
+
"loss": 0.5599,
|
| 251 |
+
"mean_token_accuracy": 0.8417543187737465,
|
| 252 |
+
"num_tokens": 3008577.0,
|
| 253 |
+
"step": 270
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.896,
|
| 257 |
+
"grad_norm": 2.2142021656036377,
|
| 258 |
+
"learning_rate": 4.779833036748801e-06,
|
| 259 |
+
"loss": 0.5268,
|
| 260 |
+
"mean_token_accuracy": 0.8514743112027645,
|
| 261 |
+
"num_tokens": 3120102.0,
|
| 262 |
+
"step": 280
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.928,
|
| 266 |
+
"grad_norm": 1.8409771919250488,
|
| 267 |
+
"learning_rate": 4.757637481550683e-06,
|
| 268 |
+
"loss": 0.5443,
|
| 269 |
+
"mean_token_accuracy": 0.8460176154971123,
|
| 270 |
+
"num_tokens": 3234456.0,
|
| 271 |
+
"step": 290
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.96,
|
| 275 |
+
"grad_norm": 1.9530022144317627,
|
| 276 |
+
"learning_rate": 4.73443290546316e-06,
|
| 277 |
+
"loss": 0.5487,
|
| 278 |
+
"mean_token_accuracy": 0.8462878614664078,
|
| 279 |
+
"num_tokens": 3348821.0,
|
| 280 |
+
"step": 300
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.992,
|
| 284 |
+
"grad_norm": 1.9719116687774658,
|
| 285 |
+
"learning_rate": 4.7102296794612865e-06,
|
| 286 |
+
"loss": 0.5229,
|
| 287 |
+
"mean_token_accuracy": 0.8531442761421204,
|
| 288 |
+
"num_tokens": 3455469.0,
|
| 289 |
+
"step": 310
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0224,
|
| 293 |
+
"grad_norm": 1.9529277086257935,
|
| 294 |
+
"learning_rate": 4.6850386208533424e-06,
|
| 295 |
+
"loss": 0.4849,
|
| 296 |
+
"mean_token_accuracy": 0.8625573249239671,
|
| 297 |
+
"num_tokens": 3561864.0,
|
| 298 |
+
"step": 320
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.0544,
|
| 302 |
+
"grad_norm": 2.105419635772705,
|
| 303 |
+
"learning_rate": 4.658870988446189e-06,
|
| 304 |
+
"loss": 0.4819,
|
| 305 |
+
"mean_token_accuracy": 0.864888047426939,
|
| 306 |
+
"num_tokens": 3672405.0,
|
| 307 |
+
"step": 330
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.0864,
|
| 311 |
+
"grad_norm": 2.035449504852295,
|
| 312 |
+
"learning_rate": 4.631738477513286e-06,
|
| 313 |
+
"loss": 0.4735,
|
| 314 |
+
"mean_token_accuracy": 0.8661263197660446,
|
| 315 |
+
"num_tokens": 3780404.0,
|
| 316 |
+
"step": 340
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.1184,
|
| 320 |
+
"grad_norm": 2.1748783588409424,
|
| 321 |
+
"learning_rate": 4.603653214567654e-06,
|
| 322 |
+
"loss": 0.4638,
|
| 323 |
+
"mean_token_accuracy": 0.8661996729671955,
|
| 324 |
+
"num_tokens": 3894359.0,
|
| 325 |
+
"step": 350
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.1504,
|
| 329 |
+
"grad_norm": 2.0794739723205566,
|
| 330 |
+
"learning_rate": 4.574627751942087e-06,
|
| 331 |
+
"loss": 0.4683,
|
| 332 |
+
"mean_token_accuracy": 0.8664901047945023,
|
| 333 |
+
"num_tokens": 4007235.0,
|
| 334 |
+
"step": 360
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.1824,
|
| 338 |
+
"grad_norm": 2.2591140270233154,
|
| 339 |
+
"learning_rate": 4.544675062179056e-06,
|
| 340 |
+
"loss": 0.4616,
|
| 341 |
+
"mean_token_accuracy": 0.8676725938916207,
|
| 342 |
+
"num_tokens": 4116127.0,
|
| 343 |
+
"step": 370
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.2144,
|
| 347 |
+
"grad_norm": 2.111358165740967,
|
| 348 |
+
"learning_rate": 4.513808532232804e-06,
|
| 349 |
+
"loss": 0.4742,
|
| 350 |
+
"mean_token_accuracy": 0.8660650610923767,
|
| 351 |
+
"num_tokens": 4228679.0,
|
| 352 |
+
"step": 380
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.2464,
|
| 356 |
+
"grad_norm": 2.115079402923584,
|
| 357 |
+
"learning_rate": 4.482041957486229e-06,
|
| 358 |
+
"loss": 0.4679,
|
| 359 |
+
"mean_token_accuracy": 0.8653517335653305,
|
| 360 |
+
"num_tokens": 4341814.0,
|
| 361 |
+
"step": 390
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.2784,
|
| 365 |
+
"grad_norm": 2.15097713470459,
|
| 366 |
+
"learning_rate": 4.4493895355852176e-06,
|
| 367 |
+
"loss": 0.448,
|
| 368 |
+
"mean_token_accuracy": 0.8716798067092896,
|
| 369 |
+
"num_tokens": 4451355.0,
|
| 370 |
+
"step": 400
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.3104,
|
| 374 |
+
"grad_norm": 2.124791383743286,
|
| 375 |
+
"learning_rate": 4.415865860093199e-06,
|
| 376 |
+
"loss": 0.4576,
|
| 377 |
+
"mean_token_accuracy": 0.8684222124516964,
|
| 378 |
+
"num_tokens": 4566924.0,
|
| 379 |
+
"step": 410
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 1.3424,
|
| 383 |
+
"grad_norm": 2.033034563064575,
|
| 384 |
+
"learning_rate": 4.381485913968747e-06,
|
| 385 |
+
"loss": 0.4558,
|
| 386 |
+
"mean_token_accuracy": 0.8672670938074589,
|
| 387 |
+
"num_tokens": 4677684.0,
|
| 388 |
+
"step": 420
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.3744,
|
| 392 |
+
"grad_norm": 2.179928779602051,
|
| 393 |
+
"learning_rate": 4.346265062869147e-06,
|
| 394 |
+
"loss": 0.468,
|
| 395 |
+
"mean_token_accuracy": 0.8666120313107968,
|
| 396 |
+
"num_tokens": 4787468.0,
|
| 397 |
+
"step": 430
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.4064,
|
| 401 |
+
"grad_norm": 2.1226675510406494,
|
| 402 |
+
"learning_rate": 4.310219048282921e-06,
|
| 403 |
+
"loss": 0.4514,
|
| 404 |
+
"mean_token_accuracy": 0.8693858951330184,
|
| 405 |
+
"num_tokens": 4899692.0,
|
| 406 |
+
"step": 440
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.4384000000000001,
|
| 410 |
+
"grad_norm": 2.226494550704956,
|
| 411 |
+
"learning_rate": 4.273363980494379e-06,
|
| 412 |
+
"loss": 0.4761,
|
| 413 |
+
"mean_token_accuracy": 0.8673883900046349,
|
| 414 |
+
"num_tokens": 5008859.0,
|
| 415 |
+
"step": 450
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.4704,
|
| 419 |
+
"grad_norm": 2.039381980895996,
|
| 420 |
+
"learning_rate": 4.235716331383343e-06,
|
| 421 |
+
"loss": 0.4652,
|
| 422 |
+
"mean_token_accuracy": 0.8683492943644524,
|
| 423 |
+
"num_tokens": 5117719.0,
|
| 424 |
+
"step": 460
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.5024,
|
| 428 |
+
"grad_norm": 2.1596784591674805,
|
| 429 |
+
"learning_rate": 4.197292927063263e-06,
|
| 430 |
+
"loss": 0.4701,
|
| 431 |
+
"mean_token_accuracy": 0.863686279207468,
|
| 432 |
+
"num_tokens": 5227041.0,
|
| 433 |
+
"step": 470
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.5344,
|
| 437 |
+
"grad_norm": 2.153256893157959,
|
| 438 |
+
"learning_rate": 4.158110940361007e-06,
|
| 439 |
+
"loss": 0.4559,
|
| 440 |
+
"mean_token_accuracy": 0.8705666847527027,
|
| 441 |
+
"num_tokens": 5334777.0,
|
| 442 |
+
"step": 480
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 1.5664,
|
| 446 |
+
"grad_norm": 2.5348920822143555,
|
| 447 |
+
"learning_rate": 4.118187883141694e-06,
|
| 448 |
+
"loss": 0.461,
|
| 449 |
+
"mean_token_accuracy": 0.8692857407033443,
|
| 450 |
+
"num_tokens": 5441237.0,
|
| 451 |
+
"step": 490
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.5984,
|
| 455 |
+
"grad_norm": 2.2534334659576416,
|
| 456 |
+
"learning_rate": 4.077541598482009e-06,
|
| 457 |
+
"loss": 0.4748,
|
| 458 |
+
"mean_token_accuracy": 0.8672291107475758,
|
| 459 |
+
"num_tokens": 5553620.0,
|
| 460 |
+
"step": 500
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.6303999999999998,
|
| 464 |
+
"grad_norm": 2.1640467643737793,
|
| 465 |
+
"learning_rate": 4.036190252695467e-06,
|
| 466 |
+
"loss": 0.4874,
|
| 467 |
+
"mean_token_accuracy": 0.8607077576220036,
|
| 468 |
+
"num_tokens": 5663373.0,
|
| 469 |
+
"step": 510
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.6623999999999999,
|
| 473 |
+
"grad_norm": 2.1866941452026367,
|
| 474 |
+
"learning_rate": 3.994152327213232e-06,
|
| 475 |
+
"loss": 0.4545,
|
| 476 |
+
"mean_token_accuracy": 0.8709075525403023,
|
| 477 |
+
"num_tokens": 5776290.0,
|
| 478 |
+
"step": 520
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.6944,
|
| 482 |
+
"grad_norm": 2.4210782051086426,
|
| 483 |
+
"learning_rate": 3.95144661032408e-06,
|
| 484 |
+
"loss": 0.4316,
|
| 485 |
+
"mean_token_accuracy": 0.8738773807883262,
|
| 486 |
+
"num_tokens": 5882282.0,
|
| 487 |
+
"step": 530
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.7264,
|
| 491 |
+
"grad_norm": 2.2260217666625977,
|
| 492 |
+
"learning_rate": 3.9080921887772314e-06,
|
| 493 |
+
"loss": 0.4497,
|
| 494 |
+
"mean_token_accuracy": 0.8696616031229496,
|
| 495 |
+
"num_tokens": 5999645.0,
|
| 496 |
+
"step": 540
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.7584,
|
| 500 |
+
"grad_norm": 2.1372995376586914,
|
| 501 |
+
"learning_rate": 3.864108439251779e-06,
|
| 502 |
+
"loss": 0.4466,
|
| 503 |
+
"mean_token_accuracy": 0.8693803034722805,
|
| 504 |
+
"num_tokens": 6108677.0,
|
| 505 |
+
"step": 550
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.7904,
|
| 509 |
+
"grad_norm": 2.0875401496887207,
|
| 510 |
+
"learning_rate": 3.8195150196965414e-06,
|
| 511 |
+
"loss": 0.4618,
|
| 512 |
+
"mean_token_accuracy": 0.866703423857689,
|
| 513 |
+
"num_tokens": 6222393.0,
|
| 514 |
+
"step": 560
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.8224,
|
| 518 |
+
"grad_norm": 2.06887149810791,
|
| 519 |
+
"learning_rate": 3.7743318605442093e-06,
|
| 520 |
+
"loss": 0.4593,
|
| 521 |
+
"mean_token_accuracy": 0.8678097896277904,
|
| 522 |
+
"num_tokens": 6337550.0,
|
| 523 |
+
"step": 570
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 1.8544,
|
| 527 |
+
"grad_norm": 1.9830808639526367,
|
| 528 |
+
"learning_rate": 3.728579155803697e-06,
|
| 529 |
+
"loss": 0.4577,
|
| 530 |
+
"mean_token_accuracy": 0.8675460621714592,
|
| 531 |
+
"num_tokens": 6449551.0,
|
| 532 |
+
"step": 580
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.8864,
|
| 536 |
+
"grad_norm": 2.165144443511963,
|
| 537 |
+
"learning_rate": 3.6822773540347107e-06,
|
| 538 |
+
"loss": 0.439,
|
| 539 |
+
"mean_token_accuracy": 0.8739401362836361,
|
| 540 |
+
"num_tokens": 6562113.0,
|
| 541 |
+
"step": 590
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.9184,
|
| 545 |
+
"grad_norm": 2.1738622188568115,
|
| 546 |
+
"learning_rate": 3.635447149208528e-06,
|
| 547 |
+
"loss": 0.4412,
|
| 548 |
+
"mean_token_accuracy": 0.8708230577409267,
|
| 549 |
+
"num_tokens": 6674860.0,
|
| 550 |
+
"step": 600
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.9504000000000001,
|
| 554 |
+
"grad_norm": 2.4379212856292725,
|
| 555 |
+
"learning_rate": 3.5881094714591102e-06,
|
| 556 |
+
"loss": 0.4522,
|
| 557 |
+
"mean_token_accuracy": 0.8691440485417843,
|
| 558 |
+
"num_tokens": 6789358.0,
|
| 559 |
+
"step": 610
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.9824000000000002,
|
| 563 |
+
"grad_norm": 2.1988134384155273,
|
| 564 |
+
"learning_rate": 3.540285477728658e-06,
|
| 565 |
+
"loss": 0.4351,
|
| 566 |
+
"mean_token_accuracy": 0.8745189763605594,
|
| 567 |
+
"num_tokens": 6899897.0,
|
| 568 |
+
"step": 620
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.0128,
|
| 572 |
+
"grad_norm": 2.226576089859009,
|
| 573 |
+
"learning_rate": 3.491996542311795e-06,
|
| 574 |
+
"loss": 0.4196,
|
| 575 |
+
"mean_token_accuracy": 0.8793611534331974,
|
| 576 |
+
"num_tokens": 7005635.0,
|
| 577 |
+
"step": 630
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 2.0448,
|
| 581 |
+
"grad_norm": 2.284040927886963,
|
| 582 |
+
"learning_rate": 3.44326424730262e-06,
|
| 583 |
+
"loss": 0.3729,
|
| 584 |
+
"mean_token_accuracy": 0.8900357685983181,
|
| 585 |
+
"num_tokens": 7118882.0,
|
| 586 |
+
"step": 640
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.0768,
|
| 590 |
+
"grad_norm": 2.3776021003723145,
|
| 591 |
+
"learning_rate": 3.394110372948871e-06,
|
| 592 |
+
"loss": 0.3784,
|
| 593 |
+
"mean_token_accuracy": 0.8912972524762154,
|
| 594 |
+
"num_tokens": 7229223.0,
|
| 595 |
+
"step": 650
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 2.1088,
|
| 599 |
+
"grad_norm": 2.3752377033233643,
|
| 600 |
+
"learning_rate": 3.344556887917539e-06,
|
| 601 |
+
"loss": 0.3754,
|
| 602 |
+
"mean_token_accuracy": 0.8933634988963604,
|
| 603 |
+
"num_tokens": 7344352.0,
|
| 604 |
+
"step": 660
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.1408,
|
| 608 |
+
"grad_norm": 2.2204787731170654,
|
| 609 |
+
"learning_rate": 3.2946259394762645e-06,
|
| 610 |
+
"loss": 0.3603,
|
| 611 |
+
"mean_token_accuracy": 0.8949612870812416,
|
| 612 |
+
"num_tokens": 7452978.0,
|
| 613 |
+
"step": 670
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.1728,
|
| 617 |
+
"grad_norm": 2.3392176628112793,
|
| 618 |
+
"learning_rate": 3.244339843594913e-06,
|
| 619 |
+
"loss": 0.3633,
|
| 620 |
+
"mean_token_accuracy": 0.8934120312333107,
|
| 621 |
+
"num_tokens": 7562987.0,
|
| 622 |
+
"step": 680
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.2048,
|
| 626 |
+
"grad_norm": 2.548868417739868,
|
| 627 |
+
"learning_rate": 3.19372107497175e-06,
|
| 628 |
+
"loss": 0.3688,
|
| 629 |
+
"mean_token_accuracy": 0.8925295993685722,
|
| 630 |
+
"num_tokens": 7672018.0,
|
| 631 |
+
"step": 690
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 2.2368,
|
| 635 |
+
"grad_norm": 2.524747371673584,
|
| 636 |
+
"learning_rate": 3.1427922569886756e-06,
|
| 637 |
+
"loss": 0.3535,
|
| 638 |
+
"mean_token_accuracy": 0.8959250211715698,
|
| 639 |
+
"num_tokens": 7778694.0,
|
| 640 |
+
"step": 700
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.2688,
|
| 644 |
+
"grad_norm": 2.2159647941589355,
|
| 645 |
+
"learning_rate": 3.0915761516000053e-06,
|
| 646 |
+
"loss": 0.3645,
|
| 647 |
+
"mean_token_accuracy": 0.8916135348379612,
|
| 648 |
+
"num_tokens": 7889175.0,
|
| 649 |
+
"step": 710
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 2.3008,
|
| 653 |
+
"grad_norm": 2.6435368061065674,
|
| 654 |
+
"learning_rate": 3.0400956491593215e-06,
|
| 655 |
+
"loss": 0.3742,
|
| 656 |
+
"mean_token_accuracy": 0.8912425316870213,
|
| 657 |
+
"num_tokens": 7997644.0,
|
| 658 |
+
"step": 720
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 2.3327999999999998,
|
| 662 |
+
"grad_norm": 2.5125110149383545,
|
| 663 |
+
"learning_rate": 2.9883737581889315e-06,
|
| 664 |
+
"loss": 0.3767,
|
| 665 |
+
"mean_token_accuracy": 0.8903645075857639,
|
| 666 |
+
"num_tokens": 8109860.0,
|
| 667 |
+
"step": 730
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.3648,
|
| 671 |
+
"grad_norm": 2.482232093811035,
|
| 672 |
+
"learning_rate": 2.9364335950965227e-06,
|
| 673 |
+
"loss": 0.3531,
|
| 674 |
+
"mean_token_accuracy": 0.8950154200196266,
|
| 675 |
+
"num_tokens": 8219541.0,
|
| 676 |
+
"step": 740
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.3968,
|
| 680 |
+
"grad_norm": 2.360308885574341,
|
| 681 |
+
"learning_rate": 2.8842983738435855e-06,
|
| 682 |
+
"loss": 0.3602,
|
| 683 |
+
"mean_token_accuracy": 0.892794082313776,
|
| 684 |
+
"num_tokens": 8329765.0,
|
| 685 |
+
"step": 750
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 2.4288,
|
| 689 |
+
"grad_norm": 2.5826432704925537,
|
| 690 |
+
"learning_rate": 2.831991395570249e-06,
|
| 691 |
+
"loss": 0.364,
|
| 692 |
+
"mean_token_accuracy": 0.8930977649986744,
|
| 693 |
+
"num_tokens": 8443917.0,
|
| 694 |
+
"step": 760
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 2.4608,
|
| 698 |
+
"grad_norm": 2.588911771774292,
|
| 699 |
+
"learning_rate": 2.779536038181146e-06,
|
| 700 |
+
"loss": 0.3771,
|
| 701 |
+
"mean_token_accuracy": 0.8915269665420056,
|
| 702 |
+
"num_tokens": 8550550.0,
|
| 703 |
+
"step": 770
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.4928,
|
| 707 |
+
"grad_norm": 2.4379560947418213,
|
| 708 |
+
"learning_rate": 2.726955745896972e-06,
|
| 709 |
+
"loss": 0.3726,
|
| 710 |
+
"mean_token_accuracy": 0.8931849867105484,
|
| 711 |
+
"num_tokens": 8659752.0,
|
| 712 |
+
"step": 780
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.5248,
|
| 716 |
+
"grad_norm": 2.285378932952881,
|
| 717 |
+
"learning_rate": 2.6742740187764026e-06,
|
| 718 |
+
"loss": 0.3644,
|
| 719 |
+
"mean_token_accuracy": 0.8937488354742527,
|
| 720 |
+
"num_tokens": 8775392.0,
|
| 721 |
+
"step": 790
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.5568,
|
| 725 |
+
"grad_norm": 2.6339035034179688,
|
| 726 |
+
"learning_rate": 2.621514402213059e-06,
|
| 727 |
+
"loss": 0.3528,
|
| 728 |
+
"mean_token_accuracy": 0.8967724144458771,
|
| 729 |
+
"num_tokens": 8888499.0,
|
| 730 |
+
"step": 800
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.5888,
|
| 734 |
+
"grad_norm": 2.2561960220336914,
|
| 735 |
+
"learning_rate": 2.568700476412204e-06,
|
| 736 |
+
"loss": 0.3846,
|
| 737 |
+
"mean_token_accuracy": 0.8879048563539982,
|
| 738 |
+
"num_tokens": 9001912.0,
|
| 739 |
+
"step": 810
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.6208,
|
| 743 |
+
"grad_norm": 2.3487298488616943,
|
| 744 |
+
"learning_rate": 2.515855845851894e-06,
|
| 745 |
+
"loss": 0.3602,
|
| 746 |
+
"mean_token_accuracy": 0.8923349373042584,
|
| 747 |
+
"num_tokens": 9113468.0,
|
| 748 |
+
"step": 820
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.6528,
|
| 752 |
+
"grad_norm": 2.4302189350128174,
|
| 753 |
+
"learning_rate": 2.46300412873326e-06,
|
| 754 |
+
"loss": 0.3613,
|
| 755 |
+
"mean_token_accuracy": 0.8928801603615284,
|
| 756 |
+
"num_tokens": 9229335.0,
|
| 757 |
+
"step": 830
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 2.6848,
|
| 761 |
+
"grad_norm": 2.48608136177063,
|
| 762 |
+
"learning_rate": 2.4101689464246816e-06,
|
| 763 |
+
"loss": 0.3642,
|
| 764 |
+
"mean_token_accuracy": 0.8935355007648468,
|
| 765 |
+
"num_tokens": 9344595.0,
|
| 766 |
+
"step": 840
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.7168,
|
| 770 |
+
"grad_norm": 2.357041120529175,
|
| 771 |
+
"learning_rate": 2.3573739129045227e-06,
|
| 772 |
+
"loss": 0.3579,
|
| 773 |
+
"mean_token_accuracy": 0.8953744530677795,
|
| 774 |
+
"num_tokens": 9455015.0,
|
| 775 |
+
"step": 850
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 2.7488,
|
| 779 |
+
"grad_norm": 2.649960517883301,
|
| 780 |
+
"learning_rate": 2.3046426242071834e-06,
|
| 781 |
+
"loss": 0.3689,
|
| 782 |
+
"mean_token_accuracy": 0.8912502139806747,
|
| 783 |
+
"num_tokens": 9560806.0,
|
| 784 |
+
"step": 860
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 2.7808,
|
| 788 |
+
"grad_norm": 2.3866405487060547,
|
| 789 |
+
"learning_rate": 2.251998647877169e-06,
|
| 790 |
+
"loss": 0.364,
|
| 791 |
+
"mean_token_accuracy": 0.8922963574528694,
|
| 792 |
+
"num_tokens": 9673310.0,
|
| 793 |
+
"step": 870
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.8128,
|
| 797 |
+
"grad_norm": 2.3306539058685303,
|
| 798 |
+
"learning_rate": 2.1994655124358875e-06,
|
| 799 |
+
"loss": 0.347,
|
| 800 |
+
"mean_token_accuracy": 0.8968791022896767,
|
| 801 |
+
"num_tokens": 9787134.0,
|
| 802 |
+
"step": 880
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.8448,
|
| 806 |
+
"grad_norm": 2.3708155155181885,
|
| 807 |
+
"learning_rate": 2.1470666968658956e-06,
|
| 808 |
+
"loss": 0.3576,
|
| 809 |
+
"mean_token_accuracy": 0.8956275820732117,
|
| 810 |
+
"num_tokens": 9901311.0,
|
| 811 |
+
"step": 890
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 2.8768000000000002,
|
| 815 |
+
"grad_norm": 2.466033935546875,
|
| 816 |
+
"learning_rate": 2.0948256201172772e-06,
|
| 817 |
+
"loss": 0.3495,
|
| 818 |
+
"mean_token_accuracy": 0.8957146309316159,
|
| 819 |
+
"num_tokens": 10017025.0,
|
| 820 |
+
"step": 900
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.9088000000000003,
|
| 824 |
+
"grad_norm": 2.630037784576416,
|
| 825 |
+
"learning_rate": 2.0427656306408594e-06,
|
| 826 |
+
"loss": 0.3643,
|
| 827 |
+
"mean_token_accuracy": 0.892961747199297,
|
| 828 |
+
"num_tokens": 10131105.0,
|
| 829 |
+
"step": 910
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 2.9408,
|
| 833 |
+
"grad_norm": 2.4972527027130127,
|
| 834 |
+
"learning_rate": 1.990909995952936e-06,
|
| 835 |
+
"loss": 0.3542,
|
| 836 |
+
"mean_token_accuracy": 0.8959409989416599,
|
| 837 |
+
"num_tokens": 10240555.0,
|
| 838 |
+
"step": 920
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 2.9728,
|
| 842 |
+
"grad_norm": 2.722083330154419,
|
| 843 |
+
"learning_rate": 1.939281892236157e-06,
|
| 844 |
+
"loss": 0.3531,
|
| 845 |
+
"mean_token_accuracy": 0.8945782586932183,
|
| 846 |
+
"num_tokens": 10347889.0,
|
| 847 |
+
"step": 930
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 3.0032,
|
| 851 |
+
"grad_norm": 2.2683236598968506,
|
| 852 |
+
"learning_rate": 1.8879043939812503e-06,
|
| 853 |
+
"loss": 0.3579,
|
| 854 |
+
"mean_token_accuracy": 0.8970388119157992,
|
| 855 |
+
"num_tokens": 10453648.0,
|
| 856 |
+
"step": 940
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 3.0352,
|
| 860 |
+
"grad_norm": 2.5510222911834717,
|
| 861 |
+
"learning_rate": 1.8368004636741794e-06,
|
| 862 |
+
"loss": 0.3062,
|
| 863 |
+
"mean_token_accuracy": 0.9120431691408157,
|
| 864 |
+
"num_tokens": 10565156.0,
|
| 865 |
+
"step": 950
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 3.0672,
|
| 869 |
+
"grad_norm": 2.730621337890625,
|
| 870 |
+
"learning_rate": 1.7859929415333725e-06,
|
| 871 |
+
"loss": 0.2912,
|
| 872 |
+
"mean_token_accuracy": 0.9148132555186749,
|
| 873 |
+
"num_tokens": 10676677.0,
|
| 874 |
+
"step": 960
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 3.0992,
|
| 878 |
+
"grad_norm": 2.5102102756500244,
|
| 879 |
+
"learning_rate": 1.735504535301592e-06,
|
| 880 |
+
"loss": 0.2882,
|
| 881 |
+
"mean_token_accuracy": 0.9159041397273541,
|
| 882 |
+
"num_tokens": 10787468.0,
|
| 883 |
+
"step": 970
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 3.1312,
|
| 887 |
+
"grad_norm": 2.3301403522491455,
|
| 888 |
+
"learning_rate": 1.6853578100970086e-06,
|
| 889 |
+
"loss": 0.3016,
|
| 890 |
+
"mean_token_accuracy": 0.911161607503891,
|
| 891 |
+
"num_tokens": 10897279.0,
|
| 892 |
+
"step": 980
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 3.1632,
|
| 896 |
+
"grad_norm": 2.4964194297790527,
|
| 897 |
+
"learning_rate": 1.6355751783280283e-06,
|
| 898 |
+
"loss": 0.2927,
|
| 899 |
+
"mean_token_accuracy": 0.9142641559243202,
|
| 900 |
+
"num_tokens": 11008731.0,
|
| 901 |
+
"step": 990
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 3.1952,
|
| 905 |
+
"grad_norm": 2.547372341156006,
|
| 906 |
+
"learning_rate": 1.5861788896763625e-06,
|
| 907 |
+
"loss": 0.2991,
|
| 908 |
+
"mean_token_accuracy": 0.9131025090813637,
|
| 909 |
+
"num_tokens": 11123820.0,
|
| 910 |
+
"step": 1000
|
| 911 |
+
}
|
| 912 |
+
],
|
| 913 |
+
"logging_steps": 10,
|
| 914 |
+
"max_steps": 1565,
|
| 915 |
+
"num_input_tokens_seen": 0,
|
| 916 |
+
"num_train_epochs": 5,
|
| 917 |
+
"save_steps": 500,
|
| 918 |
+
"stateful_callbacks": {
|
| 919 |
+
"TrainerControl": {
|
| 920 |
+
"args": {
|
| 921 |
+
"should_epoch_stop": false,
|
| 922 |
+
"should_evaluate": false,
|
| 923 |
+
"should_log": false,
|
| 924 |
+
"should_save": true,
|
| 925 |
+
"should_training_stop": false
|
| 926 |
+
},
|
| 927 |
+
"attributes": {}
|
| 928 |
+
}
|
| 929 |
+
},
|
| 930 |
+
"total_flos": 1.0242157166877082e+17,
|
| 931 |
+
"train_batch_size": 4,
|
| 932 |
+
"trial_name": null,
|
| 933 |
+
"trial_params": null
|
| 934 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:805391ab265a423fa4bf8e6272db7bca4771272624226f8273e5a50efd4ae604
|
| 3 |
+
size 6161
|