Commit
•
664c53d
1
Parent(s):
61898c1
Model save
Browse files- README.md +70 -0
- added_tokens.json +5 -0
- all_results.json +14 -0
- config.json +29 -0
- eval_results.json +8 -0
- generation_config.json +7 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +21 -0
- tokenizer.json +0 -0
- tokenizer_config.json +43 -0
- train_results.json +9 -0
- trainer_state.json +1814 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: Qwen/Qwen2-0.5B
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- alignment-handbook
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: qwen2-0.5b-sft
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/mzyin-university-of-florida/huggingface/runs/eizi1rvf)
|
| 20 |
+
# qwen2-0.5b-sft
|
| 21 |
+
|
| 22 |
+
This model is a fine-tuned version of [Qwen/Qwen2-0.5B](https://huggingface.co/Qwen/Qwen2-0.5B) on the generator dataset.
|
| 23 |
+
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 1.5327
|
| 25 |
+
|
| 26 |
+
## Model description
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Intended uses & limitations
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training and evaluation data
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training procedure
|
| 39 |
+
|
| 40 |
+
### Training hyperparameters
|
| 41 |
+
|
| 42 |
+
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 2e-05
|
| 44 |
+
- train_batch_size: 8
|
| 45 |
+
- eval_batch_size: 8
|
| 46 |
+
- seed: 42
|
| 47 |
+
- distributed_type: multi-GPU
|
| 48 |
+
- num_devices: 3
|
| 49 |
+
- gradient_accumulation_steps: 8
|
| 50 |
+
- total_train_batch_size: 192
|
| 51 |
+
- total_eval_batch_size: 24
|
| 52 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
+
- lr_scheduler_type: cosine
|
| 54 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 55 |
+
- num_epochs: 1
|
| 56 |
+
- mixed_precision_training: Native AMP
|
| 57 |
+
|
| 58 |
+
### Training results
|
| 59 |
+
|
| 60 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 61 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 62 |
+
| 1.526 | 0.9993 | 1258 | 1.5327 |
|
| 63 |
+
|
| 64 |
+
|
| 65 |
+
### Framework versions
|
| 66 |
+
|
| 67 |
+
- Transformers 4.42.0
|
| 68 |
+
- Pytorch 2.3.0+cu121
|
| 69 |
+
- Datasets 2.21.0
|
| 70 |
+
- Tokenizers 0.19.1
|
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9993049349617714,
|
| 3 |
+
"eval_loss": 1.5326406955718994,
|
| 4 |
+
"eval_runtime": 222.9042,
|
| 5 |
+
"eval_samples": 23109,
|
| 6 |
+
"eval_samples_per_second": 119.98,
|
| 7 |
+
"eval_steps_per_second": 5.002,
|
| 8 |
+
"total_flos": 106161864966144.0,
|
| 9 |
+
"train_loss": 1.5477431688475496,
|
| 10 |
+
"train_runtime": 10888.6684,
|
| 11 |
+
"train_samples": 207864,
|
| 12 |
+
"train_samples_per_second": 22.196,
|
| 13 |
+
"train_steps_per_second": 0.116
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Qwen/Qwen2-0.5B",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Qwen2ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151644,
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 896,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 4864,
|
| 13 |
+
"max_position_embeddings": 131072,
|
| 14 |
+
"max_window_layers": 24,
|
| 15 |
+
"model_type": "qwen2",
|
| 16 |
+
"num_attention_heads": 14,
|
| 17 |
+
"num_hidden_layers": 24,
|
| 18 |
+
"num_key_value_heads": 2,
|
| 19 |
+
"pad_token_id": 151645,
|
| 20 |
+
"rms_norm_eps": 1e-06,
|
| 21 |
+
"rope_theta": 1000000.0,
|
| 22 |
+
"sliding_window": 131072,
|
| 23 |
+
"tie_word_embeddings": true,
|
| 24 |
+
"torch_dtype": "float16",
|
| 25 |
+
"transformers_version": "4.42.0",
|
| 26 |
+
"use_cache": false,
|
| 27 |
+
"use_sliding_window": false,
|
| 28 |
+
"vocab_size": 151646
|
| 29 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9993049349617714,
|
| 3 |
+
"eval_loss": 1.5326406955718994,
|
| 4 |
+
"eval_runtime": 222.9042,
|
| 5 |
+
"eval_samples": 23109,
|
| 6 |
+
"eval_samples_per_second": 119.98,
|
| 7 |
+
"eval_steps_per_second": 5.002
|
| 8 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151644,
|
| 3 |
+
"eos_token_id": 151645,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"pad_token_id": 151645,
|
| 6 |
+
"transformers_version": "4.42.0"
|
| 7 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64fded441dd08833ed6f9e8a9fd34c53e165334909ebca5924e6e406eb390598
|
| 3 |
+
size 987577856
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
{
|
| 4 |
+
"content": "<|im_start|>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"content": "<|im_end|>",
|
| 12 |
+
"lstrip": false,
|
| 13 |
+
"normalized": false,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
}
|
| 17 |
+
],
|
| 18 |
+
"bos_token": "<|im_start|>",
|
| 19 |
+
"eos_token": "<|im_end|>",
|
| 20 |
+
"pad_token": "<|im_end|>"
|
| 21 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"additional_special_tokens": [
|
| 30 |
+
"<|im_start|>",
|
| 31 |
+
"<|im_end|>"
|
| 32 |
+
],
|
| 33 |
+
"bos_token": "<|im_start|>",
|
| 34 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 35 |
+
"clean_up_tokenization_spaces": false,
|
| 36 |
+
"eos_token": "<|im_end|>",
|
| 37 |
+
"errors": "replace",
|
| 38 |
+
"model_max_length": 32768,
|
| 39 |
+
"pad_token": "<|im_end|>",
|
| 40 |
+
"split_special_tokens": false,
|
| 41 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 42 |
+
"unk_token": null
|
| 43 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9993049349617714,
|
| 3 |
+
"total_flos": 106161864966144.0,
|
| 4 |
+
"train_loss": 1.5477431688475496,
|
| 5 |
+
"train_runtime": 10888.6684,
|
| 6 |
+
"train_samples": 207864,
|
| 7 |
+
"train_samples_per_second": 22.196,
|
| 8 |
+
"train_steps_per_second": 0.116
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1814 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9993049349617714,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1258,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0007943600436898023,
|
| 13 |
+
"grad_norm": 11.12145892458997,
|
| 14 |
+
"learning_rate": 1.5873015873015874e-07,
|
| 15 |
+
"loss": 1.8009,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003971800218449012,
|
| 20 |
+
"grad_norm": 10.234395651621842,
|
| 21 |
+
"learning_rate": 7.936507936507937e-07,
|
| 22 |
+
"loss": 1.7723,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.007943600436898023,
|
| 27 |
+
"grad_norm": 2.755899780672782,
|
| 28 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 29 |
+
"loss": 1.709,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.011915400655347037,
|
| 34 |
+
"grad_norm": 1.5712351118319727,
|
| 35 |
+
"learning_rate": 2.380952380952381e-06,
|
| 36 |
+
"loss": 1.6524,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.015887200873796047,
|
| 41 |
+
"grad_norm": 1.1660528045797167,
|
| 42 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 43 |
+
"loss": 1.6423,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01985900109224506,
|
| 48 |
+
"grad_norm": 1.0216120526399777,
|
| 49 |
+
"learning_rate": 3.968253968253968e-06,
|
| 50 |
+
"loss": 1.6244,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.023830801310694073,
|
| 55 |
+
"grad_norm": 0.9509263430769295,
|
| 56 |
+
"learning_rate": 4.761904761904762e-06,
|
| 57 |
+
"loss": 1.6241,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.027802601529143083,
|
| 62 |
+
"grad_norm": 0.9135838434062775,
|
| 63 |
+
"learning_rate": 5.555555555555557e-06,
|
| 64 |
+
"loss": 1.6066,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03177440174759209,
|
| 69 |
+
"grad_norm": 0.9277594740139052,
|
| 70 |
+
"learning_rate": 6.349206349206349e-06,
|
| 71 |
+
"loss": 1.5847,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.035746201966041107,
|
| 76 |
+
"grad_norm": 0.8992831827632032,
|
| 77 |
+
"learning_rate": 7.1428571428571436e-06,
|
| 78 |
+
"loss": 1.5999,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03971800218449012,
|
| 83 |
+
"grad_norm": 0.9718735971756692,
|
| 84 |
+
"learning_rate": 7.936507936507936e-06,
|
| 85 |
+
"loss": 1.6023,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04368980240293913,
|
| 90 |
+
"grad_norm": 0.947148151716203,
|
| 91 |
+
"learning_rate": 8.730158730158731e-06,
|
| 92 |
+
"loss": 1.5825,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04766160262138815,
|
| 97 |
+
"grad_norm": 0.9159573540177632,
|
| 98 |
+
"learning_rate": 9.523809523809525e-06,
|
| 99 |
+
"loss": 1.5978,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05163340283983715,
|
| 104 |
+
"grad_norm": 0.9226608146834196,
|
| 105 |
+
"learning_rate": 1.031746031746032e-05,
|
| 106 |
+
"loss": 1.6003,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.055605203058286166,
|
| 111 |
+
"grad_norm": 0.9212318659674721,
|
| 112 |
+
"learning_rate": 1.1111111111111113e-05,
|
| 113 |
+
"loss": 1.5856,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05957700327673518,
|
| 118 |
+
"grad_norm": 0.8933451806059763,
|
| 119 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 120 |
+
"loss": 1.5845,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06354880349518419,
|
| 125 |
+
"grad_norm": 0.9657915647359273,
|
| 126 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 127 |
+
"loss": 1.6089,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0675206037136332,
|
| 132 |
+
"grad_norm": 0.9337955988860143,
|
| 133 |
+
"learning_rate": 1.3492063492063494e-05,
|
| 134 |
+
"loss": 1.5815,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07149240393208221,
|
| 139 |
+
"grad_norm": 0.9873409918119863,
|
| 140 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 141 |
+
"loss": 1.5816,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07546420415053123,
|
| 146 |
+
"grad_norm": 0.9015172917838395,
|
| 147 |
+
"learning_rate": 1.507936507936508e-05,
|
| 148 |
+
"loss": 1.5958,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07943600436898024,
|
| 153 |
+
"grad_norm": 0.9772975467717536,
|
| 154 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 155 |
+
"loss": 1.5911,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08340780458742925,
|
| 160 |
+
"grad_norm": 0.9472359700179834,
|
| 161 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 162 |
+
"loss": 1.5865,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08737960480587827,
|
| 167 |
+
"grad_norm": 0.9247784591978276,
|
| 168 |
+
"learning_rate": 1.7460317460317463e-05,
|
| 169 |
+
"loss": 1.5793,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09135140502432727,
|
| 174 |
+
"grad_norm": 0.9231839444533193,
|
| 175 |
+
"learning_rate": 1.8253968253968254e-05,
|
| 176 |
+
"loss": 1.5822,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0953232052427763,
|
| 181 |
+
"grad_norm": 0.9573224627010042,
|
| 182 |
+
"learning_rate": 1.904761904761905e-05,
|
| 183 |
+
"loss": 1.5718,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0992950054612253,
|
| 188 |
+
"grad_norm": 1.0158728633269807,
|
| 189 |
+
"learning_rate": 1.9841269841269845e-05,
|
| 190 |
+
"loss": 1.5773,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.1032668056796743,
|
| 195 |
+
"grad_norm": 1.0442473658793399,
|
| 196 |
+
"learning_rate": 1.999938384153589e-05,
|
| 197 |
+
"loss": 1.585,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10723860589812333,
|
| 202 |
+
"grad_norm": 0.9568516629501055,
|
| 203 |
+
"learning_rate": 1.999688082790923e-05,
|
| 204 |
+
"loss": 1.5868,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11121040611657233,
|
| 209 |
+
"grad_norm": 0.9665579055787764,
|
| 210 |
+
"learning_rate": 1.9992452930796544e-05,
|
| 211 |
+
"loss": 1.5776,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11518220633502135,
|
| 216 |
+
"grad_norm": 0.9323962932240065,
|
| 217 |
+
"learning_rate": 1.9986101002782376e-05,
|
| 218 |
+
"loss": 1.5789,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11915400655347036,
|
| 223 |
+
"grad_norm": 0.9477001351382732,
|
| 224 |
+
"learning_rate": 1.997782626692034e-05,
|
| 225 |
+
"loss": 1.5814,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.12312580677191937,
|
| 230 |
+
"grad_norm": 0.9311645028761003,
|
| 231 |
+
"learning_rate": 1.9967630316497663e-05,
|
| 232 |
+
"loss": 1.5658,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12709760699036837,
|
| 237 |
+
"grad_norm": 0.9585549328757471,
|
| 238 |
+
"learning_rate": 1.995551511472836e-05,
|
| 239 |
+
"loss": 1.5843,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.1310694072088174,
|
| 244 |
+
"grad_norm": 0.94350147619133,
|
| 245 |
+
"learning_rate": 1.994148299437524e-05,
|
| 246 |
+
"loss": 1.5589,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1350412074272664,
|
| 251 |
+
"grad_norm": 0.9702241501746585,
|
| 252 |
+
"learning_rate": 1.9925536657300734e-05,
|
| 253 |
+
"loss": 1.5783,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.13901300764571542,
|
| 258 |
+
"grad_norm": 0.90878186459247,
|
| 259 |
+
"learning_rate": 1.990767917394666e-05,
|
| 260 |
+
"loss": 1.5716,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.14298480786416443,
|
| 265 |
+
"grad_norm": 0.9926126694962801,
|
| 266 |
+
"learning_rate": 1.9887913982743e-05,
|
| 267 |
+
"loss": 1.5705,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.14695660808261343,
|
| 272 |
+
"grad_norm": 0.8958939695984969,
|
| 273 |
+
"learning_rate": 1.986624488944585e-05,
|
| 274 |
+
"loss": 1.5738,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.15092840830106247,
|
| 279 |
+
"grad_norm": 0.9194318242066775,
|
| 280 |
+
"learning_rate": 1.984267606640462e-05,
|
| 281 |
+
"loss": 1.5729,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.15490020851951147,
|
| 286 |
+
"grad_norm": 0.9497866986358652,
|
| 287 |
+
"learning_rate": 1.9817212051758667e-05,
|
| 288 |
+
"loss": 1.5674,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.15887200873796048,
|
| 293 |
+
"grad_norm": 0.8988871329419085,
|
| 294 |
+
"learning_rate": 1.978985774856346e-05,
|
| 295 |
+
"loss": 1.5683,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.16284380895640949,
|
| 300 |
+
"grad_norm": 0.9722584116447944,
|
| 301 |
+
"learning_rate": 1.9760618423846526e-05,
|
| 302 |
+
"loss": 1.5737,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1668156091748585,
|
| 307 |
+
"grad_norm": 0.9803590939025236,
|
| 308 |
+
"learning_rate": 1.9729499707593284e-05,
|
| 309 |
+
"loss": 1.5826,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.17078740939330753,
|
| 314 |
+
"grad_norm": 0.9344503381295605,
|
| 315 |
+
"learning_rate": 1.9696507591663003e-05,
|
| 316 |
+
"loss": 1.5565,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.17475920961175653,
|
| 321 |
+
"grad_norm": 0.9071381723958235,
|
| 322 |
+
"learning_rate": 1.9661648428635066e-05,
|
| 323 |
+
"loss": 1.5621,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.17873100983020554,
|
| 328 |
+
"grad_norm": 0.9288351049804889,
|
| 329 |
+
"learning_rate": 1.962492893058582e-05,
|
| 330 |
+
"loss": 1.5532,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.18270281004865455,
|
| 335 |
+
"grad_norm": 0.9288525670728277,
|
| 336 |
+
"learning_rate": 1.9586356167796145e-05,
|
| 337 |
+
"loss": 1.5801,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.18667461026710355,
|
| 342 |
+
"grad_norm": 0.9183899709915191,
|
| 343 |
+
"learning_rate": 1.954593756739009e-05,
|
| 344 |
+
"loss": 1.5801,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1906464104855526,
|
| 349 |
+
"grad_norm": 0.9877621229108748,
|
| 350 |
+
"learning_rate": 1.9512278901942467e-05,
|
| 351 |
+
"loss": 1.5817,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.1946182107040016,
|
| 356 |
+
"grad_norm": 0.9364533708770005,
|
| 357 |
+
"learning_rate": 1.9468557643703262e-05,
|
| 358 |
+
"loss": 1.571,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.1985900109224506,
|
| 363 |
+
"grad_norm": 0.9156871855852468,
|
| 364 |
+
"learning_rate": 1.942301322976593e-05,
|
| 365 |
+
"loss": 1.5693,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2025618111408996,
|
| 370 |
+
"grad_norm": 0.91239458392149,
|
| 371 |
+
"learning_rate": 1.9375654429634866e-05,
|
| 372 |
+
"loss": 1.556,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2065336113593486,
|
| 377 |
+
"grad_norm": 0.904505667076919,
|
| 378 |
+
"learning_rate": 1.9326490362171625e-05,
|
| 379 |
+
"loss": 1.5763,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.21050541157779765,
|
| 384 |
+
"grad_norm": 0.901682231936739,
|
| 385 |
+
"learning_rate": 1.9275530493839118e-05,
|
| 386 |
+
"loss": 1.5706,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.21447721179624665,
|
| 391 |
+
"grad_norm": 0.9063579769395875,
|
| 392 |
+
"learning_rate": 1.9222784636878853e-05,
|
| 393 |
+
"loss": 1.567,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.21844901201469566,
|
| 398 |
+
"grad_norm": 0.9214787619294059,
|
| 399 |
+
"learning_rate": 1.91682629474216e-05,
|
| 400 |
+
"loss": 1.5742,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.22242081223314467,
|
| 405 |
+
"grad_norm": 0.9524031793810314,
|
| 406 |
+
"learning_rate": 1.9111975923531858e-05,
|
| 407 |
+
"loss": 1.5653,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.22639261245159367,
|
| 412 |
+
"grad_norm": 0.9437628509942092,
|
| 413 |
+
"learning_rate": 1.905393440318645e-05,
|
| 414 |
+
"loss": 1.555,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2303644126700427,
|
| 419 |
+
"grad_norm": 0.9556933869893008,
|
| 420 |
+
"learning_rate": 1.8994149562187702e-05,
|
| 421 |
+
"loss": 1.5625,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2343362128884917,
|
| 426 |
+
"grad_norm": 0.9430484585606292,
|
| 427 |
+
"learning_rate": 1.8932632912011565e-05,
|
| 428 |
+
"loss": 1.5567,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.23830801310694072,
|
| 433 |
+
"grad_norm": 0.9741243595635902,
|
| 434 |
+
"learning_rate": 1.886939629759107e-05,
|
| 435 |
+
"loss": 1.5553,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.24227981332538973,
|
| 440 |
+
"grad_norm": 0.9052975360645378,
|
| 441 |
+
"learning_rate": 1.8804451895035645e-05,
|
| 442 |
+
"loss": 1.5793,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.24625161354383873,
|
| 447 |
+
"grad_norm": 0.9742325579737205,
|
| 448 |
+
"learning_rate": 1.873781220928659e-05,
|
| 449 |
+
"loss": 1.5605,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.25022341376228774,
|
| 454 |
+
"grad_norm": 0.9031213199856675,
|
| 455 |
+
"learning_rate": 1.866949007170929e-05,
|
| 456 |
+
"loss": 1.5649,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.25419521398073675,
|
| 461 |
+
"grad_norm": 0.8702720949465248,
|
| 462 |
+
"learning_rate": 1.859949863762256e-05,
|
| 463 |
+
"loss": 1.5749,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2581670141991858,
|
| 468 |
+
"grad_norm": 0.961271121572224,
|
| 469 |
+
"learning_rate": 1.852785138376558e-05,
|
| 470 |
+
"loss": 1.5752,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2621388144176348,
|
| 475 |
+
"grad_norm": 0.8718585392590509,
|
| 476 |
+
"learning_rate": 1.8454562105703e-05,
|
| 477 |
+
"loss": 1.5577,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2661106146360838,
|
| 482 |
+
"grad_norm": 0.8690540556968002,
|
| 483 |
+
"learning_rate": 1.8379644915168623e-05,
|
| 484 |
+
"loss": 1.55,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2700824148545328,
|
| 489 |
+
"grad_norm": 0.9097212286930987,
|
| 490 |
+
"learning_rate": 1.83031142373482e-05,
|
| 491 |
+
"loss": 1.5737,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.27405421507298183,
|
| 496 |
+
"grad_norm": 0.8712230897009824,
|
| 497 |
+
"learning_rate": 1.822498480810189e-05,
|
| 498 |
+
"loss": 1.5739,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.27802601529143084,
|
| 503 |
+
"grad_norm": 0.948552271738661,
|
| 504 |
+
"learning_rate": 1.8145271671126892e-05,
|
| 505 |
+
"loss": 1.5657,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.28199781550987985,
|
| 510 |
+
"grad_norm": 0.9284269691863282,
|
| 511 |
+
"learning_rate": 1.8063990175060807e-05,
|
| 512 |
+
"loss": 1.5601,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.28596961572832885,
|
| 517 |
+
"grad_norm": 0.8854698701217475,
|
| 518 |
+
"learning_rate": 1.798115597052629e-05,
|
| 519 |
+
"loss": 1.5595,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.28994141594677786,
|
| 524 |
+
"grad_norm": 0.9222629630617156,
|
| 525 |
+
"learning_rate": 1.7896785007117526e-05,
|
| 526 |
+
"loss": 1.5682,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.29391321616522686,
|
| 531 |
+
"grad_norm": 0.9259111426870978,
|
| 532 |
+
"learning_rate": 1.781089353032918e-05,
|
| 533 |
+
"loss": 1.5614,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2978850163836759,
|
| 538 |
+
"grad_norm": 0.9425926440581255,
|
| 539 |
+
"learning_rate": 1.7723498078428355e-05,
|
| 540 |
+
"loss": 1.5726,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.30185681660212493,
|
| 545 |
+
"grad_norm": 0.8788362482569841,
|
| 546 |
+
"learning_rate": 1.7634615479270157e-05,
|
| 547 |
+
"loss": 1.542,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.30582861682057394,
|
| 552 |
+
"grad_norm": 0.8993764823284865,
|
| 553 |
+
"learning_rate": 1.754426284705753e-05,
|
| 554 |
+
"loss": 1.5517,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.30980041703902295,
|
| 559 |
+
"grad_norm": 0.9114957468679018,
|
| 560 |
+
"learning_rate": 1.7452457579045948e-05,
|
| 561 |
+
"loss": 1.5566,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.31377221725747195,
|
| 566 |
+
"grad_norm": 0.9303896756930642,
|
| 567 |
+
"learning_rate": 1.7359217352193587e-05,
|
| 568 |
+
"loss": 1.5733,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.31774401747592096,
|
| 573 |
+
"grad_norm": 0.909449512224505,
|
| 574 |
+
"learning_rate": 1.726456011975767e-05,
|
| 575 |
+
"loss": 1.5492,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.32171581769436997,
|
| 580 |
+
"grad_norm": 0.9015167969914157,
|
| 581 |
+
"learning_rate": 1.716850410783758e-05,
|
| 582 |
+
"loss": 1.5696,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.32568761791281897,
|
| 587 |
+
"grad_norm": 0.9132981601959252,
|
| 588 |
+
"learning_rate": 1.7071067811865477e-05,
|
| 589 |
+
"loss": 1.5507,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.329659418131268,
|
| 594 |
+
"grad_norm": 0.8960449583705886,
|
| 595 |
+
"learning_rate": 1.6972269993045004e-05,
|
| 596 |
+
"loss": 1.5903,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.333631218349717,
|
| 601 |
+
"grad_norm": 0.8844544873628968,
|
| 602 |
+
"learning_rate": 1.6872129674738866e-05,
|
| 603 |
+
"loss": 1.5593,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.33760301856816605,
|
| 608 |
+
"grad_norm": 0.9059334124525383,
|
| 609 |
+
"learning_rate": 1.6770666138805904e-05,
|
| 610 |
+
"loss": 1.5829,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.34157481878661505,
|
| 615 |
+
"grad_norm": 0.8586072355217647,
|
| 616 |
+
"learning_rate": 1.666789892188841e-05,
|
| 617 |
+
"loss": 1.5577,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.34554661900506406,
|
| 622 |
+
"grad_norm": 0.929783164434847,
|
| 623 |
+
"learning_rate": 1.6563847811650376e-05,
|
| 624 |
+
"loss": 1.5684,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.34951841922351307,
|
| 629 |
+
"grad_norm": 0.8725442050763763,
|
| 630 |
+
"learning_rate": 1.64585328429674e-05,
|
| 631 |
+
"loss": 1.5448,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3534902194419621,
|
| 636 |
+
"grad_norm": 0.8749100355256942,
|
| 637 |
+
"learning_rate": 1.635197429406901e-05,
|
| 638 |
+
"loss": 1.5726,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3574620196604111,
|
| 643 |
+
"grad_norm": 0.948191551262984,
|
| 644 |
+
"learning_rate": 1.6244192682634143e-05,
|
| 645 |
+
"loss": 1.5464,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3614338198788601,
|
| 650 |
+
"grad_norm": 0.9305834084339885,
|
| 651 |
+
"learning_rate": 1.6135208761840457e-05,
|
| 652 |
+
"loss": 1.559,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3654056200973091,
|
| 657 |
+
"grad_norm": 0.9397991327586414,
|
| 658 |
+
"learning_rate": 1.602504351636838e-05,
|
| 659 |
+
"loss": 1.5534,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3693774203157581,
|
| 664 |
+
"grad_norm": 0.9213324585480839,
|
| 665 |
+
"learning_rate": 1.591371815836051e-05,
|
| 666 |
+
"loss": 1.5542,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3733492205342071,
|
| 671 |
+
"grad_norm": 0.8853761110334394,
|
| 672 |
+
"learning_rate": 1.580125412333728e-05,
|
| 673 |
+
"loss": 1.5402,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.37732102075265617,
|
| 678 |
+
"grad_norm": 0.8691498605561269,
|
| 679 |
+
"learning_rate": 1.5687673066069568e-05,
|
| 680 |
+
"loss": 1.552,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3812928209711052,
|
| 685 |
+
"grad_norm": 0.8998572635366405,
|
| 686 |
+
"learning_rate": 1.5572996856409094e-05,
|
| 687 |
+
"loss": 1.5638,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3852646211895542,
|
| 692 |
+
"grad_norm": 0.8941950276870779,
|
| 693 |
+
"learning_rate": 1.5457247575077445e-05,
|
| 694 |
+
"loss": 1.5406,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3892364214080032,
|
| 699 |
+
"grad_norm": 0.9097426424762942,
|
| 700 |
+
"learning_rate": 1.534044750941444e-05,
|
| 701 |
+
"loss": 1.5472,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3932082216264522,
|
| 706 |
+
"grad_norm": 0.8791345662658966,
|
| 707 |
+
"learning_rate": 1.5222619149086746e-05,
|
| 708 |
+
"loss": 1.5413,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3971800218449012,
|
| 713 |
+
"grad_norm": 0.8776919953132802,
|
| 714 |
+
"learning_rate": 1.5103785181757533e-05,
|
| 715 |
+
"loss": 1.5396,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.4011518220633502,
|
| 720 |
+
"grad_norm": 0.8664994594605832,
|
| 721 |
+
"learning_rate": 1.4983968488718005e-05,
|
| 722 |
+
"loss": 1.5426,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.4051236222817992,
|
| 727 |
+
"grad_norm": 0.8931546183056348,
|
| 728 |
+
"learning_rate": 1.4863192140481624e-05,
|
| 729 |
+
"loss": 1.5537,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.4090954225002482,
|
| 734 |
+
"grad_norm": 0.8442200510246032,
|
| 735 |
+
"learning_rate": 1.4741479392341941e-05,
|
| 736 |
+
"loss": 1.5586,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.4130672227186972,
|
| 741 |
+
"grad_norm": 0.9413240109314772,
|
| 742 |
+
"learning_rate": 1.4618853679894813e-05,
|
| 743 |
+
"loss": 1.5202,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.4170390229371463,
|
| 748 |
+
"grad_norm": 0.8865446672373302,
|
| 749 |
+
"learning_rate": 1.4495338614525927e-05,
|
| 750 |
+
"loss": 1.5503,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.4210108231555953,
|
| 755 |
+
"grad_norm": 0.9668394693388267,
|
| 756 |
+
"learning_rate": 1.437095797886445e-05,
|
| 757 |
+
"loss": 1.5488,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.4249826233740443,
|
| 762 |
+
"grad_norm": 0.9214606309227755,
|
| 763 |
+
"learning_rate": 1.4245735722203736e-05,
|
| 764 |
+
"loss": 1.54,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4289544235924933,
|
| 769 |
+
"grad_norm": 0.9555320905995948,
|
| 770 |
+
"learning_rate": 1.4119695955889925e-05,
|
| 771 |
+
"loss": 1.5492,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4329262238109423,
|
| 776 |
+
"grad_norm": 0.8802626029124945,
|
| 777 |
+
"learning_rate": 1.3992862948679332e-05,
|
| 778 |
+
"loss": 1.549,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4368980240293913,
|
| 783 |
+
"grad_norm": 0.8902159443351422,
|
| 784 |
+
"learning_rate": 1.3865261122065551e-05,
|
| 785 |
+
"loss": 1.5482,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4408698242478403,
|
| 790 |
+
"grad_norm": 0.8427883741871806,
|
| 791 |
+
"learning_rate": 1.3736915045577122e-05,
|
| 792 |
+
"loss": 1.5487,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.44484162446628933,
|
| 797 |
+
"grad_norm": 0.8251277406307889,
|
| 798 |
+
"learning_rate": 1.3607849432046717e-05,
|
| 799 |
+
"loss": 1.5478,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.44881342468473834,
|
| 804 |
+
"grad_norm": 0.8371689103015839,
|
| 805 |
+
"learning_rate": 1.3478089132852717e-05,
|
| 806 |
+
"loss": 1.5597,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.45278522490318734,
|
| 811 |
+
"grad_norm": 0.8217449927915415,
|
| 812 |
+
"learning_rate": 1.3347659133134118e-05,
|
| 813 |
+
"loss": 1.5141,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4567570251216364,
|
| 818 |
+
"grad_norm": 0.8372002721470211,
|
| 819 |
+
"learning_rate": 1.3216584546979702e-05,
|
| 820 |
+
"loss": 1.5338,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4607288253400854,
|
| 825 |
+
"grad_norm": 0.9501487795845303,
|
| 826 |
+
"learning_rate": 1.3084890612592325e-05,
|
| 827 |
+
"loss": 1.5633,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4647006255585344,
|
| 832 |
+
"grad_norm": 0.8558411230061426,
|
| 833 |
+
"learning_rate": 1.2952602687429364e-05,
|
| 834 |
+
"loss": 1.5623,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4686724257769834,
|
| 839 |
+
"grad_norm": 0.830027301501606,
|
| 840 |
+
"learning_rate": 1.2819746243320176e-05,
|
| 841 |
+
"loss": 1.5512,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.47264422599543243,
|
| 846 |
+
"grad_norm": 0.903709450047426,
|
| 847 |
+
"learning_rate": 1.2686346861561538e-05,
|
| 848 |
+
"loss": 1.5608,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.47661602621388144,
|
| 853 |
+
"grad_norm": 0.961449581540646,
|
| 854 |
+
"learning_rate": 1.2552430227992005e-05,
|
| 855 |
+
"loss": 1.542,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.48058782643233044,
|
| 860 |
+
"grad_norm": 0.877769489004286,
|
| 861 |
+
"learning_rate": 1.2418022128046144e-05,
|
| 862 |
+
"loss": 1.539,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.48455962665077945,
|
| 867 |
+
"grad_norm": 0.9130539885425439,
|
| 868 |
+
"learning_rate": 1.2283148441789586e-05,
|
| 869 |
+
"loss": 1.5276,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.48853142686922846,
|
| 874 |
+
"grad_norm": 0.882980412032183,
|
| 875 |
+
"learning_rate": 1.2147835138935868e-05,
|
| 876 |
+
"loss": 1.5204,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.49250322708767746,
|
| 881 |
+
"grad_norm": 0.8366128423326633,
|
| 882 |
+
"learning_rate": 1.2012108273846011e-05,
|
| 883 |
+
"loss": 1.5362,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4964750273061265,
|
| 888 |
+
"grad_norm": 0.8538967577380823,
|
| 889 |
+
"learning_rate": 1.1875993980511772e-05,
|
| 890 |
+
"loss": 1.5187,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.5004468275245755,
|
| 895 |
+
"grad_norm": 0.8942558682507143,
|
| 896 |
+
"learning_rate": 1.1739518467523614e-05,
|
| 897 |
+
"loss": 1.5386,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5044186277430245,
|
| 902 |
+
"grad_norm": 0.9187709823388034,
|
| 903 |
+
"learning_rate": 1.1602708013024255e-05,
|
| 904 |
+
"loss": 1.5245,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5083904279614735,
|
| 909 |
+
"grad_norm": 0.8442877528307388,
|
| 910 |
+
"learning_rate": 1.146558895964888e-05,
|
| 911 |
+
"loss": 1.5596,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.5123622281799225,
|
| 916 |
+
"grad_norm": 0.898548400584606,
|
| 917 |
+
"learning_rate": 1.1328187709452884e-05,
|
| 918 |
+
"loss": 1.5406,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5163340283983716,
|
| 923 |
+
"grad_norm": 0.8811961105698817,
|
| 924 |
+
"learning_rate": 1.119053071882822e-05,
|
| 925 |
+
"loss": 1.5356,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5203058286168206,
|
| 930 |
+
"grad_norm": 0.8804203720770784,
|
| 931 |
+
"learning_rate": 1.1052644493409255e-05,
|
| 932 |
+
"loss": 1.5528,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5242776288352696,
|
| 937 |
+
"grad_norm": 0.8797158963008079,
|
| 938 |
+
"learning_rate": 1.091455558296914e-05,
|
| 939 |
+
"loss": 1.5372,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5282494290537186,
|
| 944 |
+
"grad_norm": 0.8350754715847598,
|
| 945 |
+
"learning_rate": 1.077629057630771e-05,
|
| 946 |
+
"loss": 1.5452,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5322212292721676,
|
| 951 |
+
"grad_norm": 0.8903077066951119,
|
| 952 |
+
"learning_rate": 1.0637876096131852e-05,
|
| 953 |
+
"loss": 1.5283,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5361930294906166,
|
| 958 |
+
"grad_norm": 0.8576752803021083,
|
| 959 |
+
"learning_rate": 1.049933879392933e-05,
|
| 960 |
+
"loss": 1.5433,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5401648297090657,
|
| 965 |
+
"grad_norm": 0.8467022082112134,
|
| 966 |
+
"learning_rate": 1.0360705344837105e-05,
|
| 967 |
+
"loss": 1.5133,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5441366299275147,
|
| 972 |
+
"grad_norm": 0.8817844142301194,
|
| 973 |
+
"learning_rate": 1.0222002442505057e-05,
|
| 974 |
+
"loss": 1.5411,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5481084301459637,
|
| 979 |
+
"grad_norm": 0.8151311907804857,
|
| 980 |
+
"learning_rate": 1.0083256793956186e-05,
|
| 981 |
+
"loss": 1.5326,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5520802303644127,
|
| 986 |
+
"grad_norm": 0.8251603017546312,
|
| 987 |
+
"learning_rate": 9.944495114444205e-06,
|
| 988 |
+
"loss": 1.5082,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5560520305828617,
|
| 993 |
+
"grad_norm": 0.8869207810981268,
|
| 994 |
+
"learning_rate": 9.805744122309557e-06,
|
| 995 |
+
"loss": 1.5481,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5600238308013107,
|
| 1000 |
+
"grad_norm": 0.8719890884427858,
|
| 1001 |
+
"learning_rate": 9.667030533834862e-06,
|
| 1002 |
+
"loss": 1.5223,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5639956310197597,
|
| 1007 |
+
"grad_norm": 0.8664009171759752,
|
| 1008 |
+
"learning_rate": 9.528381058100707e-06,
|
| 1009 |
+
"loss": 1.5289,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5679674312382087,
|
| 1014 |
+
"grad_norm": 0.8819223345245485,
|
| 1015 |
+
"learning_rate": 9.389822391842886e-06,
|
| 1016 |
+
"loss": 1.5248,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5719392314566577,
|
| 1021 |
+
"grad_norm": 0.8462371220192982,
|
| 1022 |
+
"learning_rate": 9.251381214311974e-06,
|
| 1023 |
+
"loss": 1.5263,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5759110316751067,
|
| 1028 |
+
"grad_norm": 0.8556000311877207,
|
| 1029 |
+
"learning_rate": 9.113084182136267e-06,
|
| 1030 |
+
"loss": 1.5464,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5798828318935557,
|
| 1035 |
+
"grad_norm": 0.8854678303214264,
|
| 1036 |
+
"learning_rate": 8.974957924189108e-06,
|
| 1037 |
+
"loss": 1.5388,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5838546321120047,
|
| 1042 |
+
"grad_norm": 0.8605998763470911,
|
| 1043 |
+
"learning_rate": 8.837029036461537e-06,
|
| 1044 |
+
"loss": 1.5391,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5878264323304537,
|
| 1049 |
+
"grad_norm": 0.8176529738699428,
|
| 1050 |
+
"learning_rate": 8.699324076941278e-06,
|
| 1051 |
+
"loss": 1.5578,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5917982325489027,
|
| 1056 |
+
"grad_norm": 0.8500747154468798,
|
| 1057 |
+
"learning_rate": 8.561869560499017e-06,
|
| 1058 |
+
"loss": 1.5387,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5957700327673519,
|
| 1063 |
+
"grad_norm": 0.8074215452026124,
|
| 1064 |
+
"learning_rate": 8.42469195378304e-06,
|
| 1065 |
+
"loss": 1.5312,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5997418329858009,
|
| 1070 |
+
"grad_norm": 0.8452784103185206,
|
| 1071 |
+
"learning_rate": 8.287817670123101e-06,
|
| 1072 |
+
"loss": 1.5258,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.6037136332042499,
|
| 1077 |
+
"grad_norm": 0.8486586156355476,
|
| 1078 |
+
"learning_rate": 8.151273064444583e-06,
|
| 1079 |
+
"loss": 1.5192,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6076854334226989,
|
| 1084 |
+
"grad_norm": 0.8146588435190393,
|
| 1085 |
+
"learning_rate": 8.015084428193895e-06,
|
| 1086 |
+
"loss": 1.5045,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.6116572336411479,
|
| 1091 |
+
"grad_norm": 0.8320002856800587,
|
| 1092 |
+
"learning_rate": 7.879277984276106e-06,
|
| 1093 |
+
"loss": 1.5371,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.6156290338595969,
|
| 1098 |
+
"grad_norm": 0.8314805091840901,
|
| 1099 |
+
"learning_rate": 7.74387988200576e-06,
|
| 1100 |
+
"loss": 1.5364,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6196008340780459,
|
| 1105 |
+
"grad_norm": 0.8340274339801351,
|
| 1106 |
+
"learning_rate": 7.608916192071856e-06,
|
| 1107 |
+
"loss": 1.5326,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6235726342964949,
|
| 1112 |
+
"grad_norm": 0.8445705617901014,
|
| 1113 |
+
"learning_rate": 7.474412901517998e-06,
|
| 1114 |
+
"loss": 1.543,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6275444345149439,
|
| 1119 |
+
"grad_norm": 0.8631402798545373,
|
| 1120 |
+
"learning_rate": 7.340395908738622e-06,
|
| 1121 |
+
"loss": 1.5258,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6315162347333929,
|
| 1126 |
+
"grad_norm": 0.839562146660397,
|
| 1127 |
+
"learning_rate": 7.206891018492308e-06,
|
| 1128 |
+
"loss": 1.5464,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6354880349518419,
|
| 1133 |
+
"grad_norm": 0.879806527044698,
|
| 1134 |
+
"learning_rate": 7.073923936933091e-06,
|
| 1135 |
+
"loss": 1.541,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6394598351702909,
|
| 1140 |
+
"grad_norm": 0.8536606816656509,
|
| 1141 |
+
"learning_rate": 6.941520266660819e-06,
|
| 1142 |
+
"loss": 1.5206,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6434316353887399,
|
| 1147 |
+
"grad_norm": 0.8318733759254066,
|
| 1148 |
+
"learning_rate": 6.809705501791379e-06,
|
| 1149 |
+
"loss": 1.5407,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.6474034356071889,
|
| 1154 |
+
"grad_norm": 0.8459590871404428,
|
| 1155 |
+
"learning_rate": 6.678505023047871e-06,
|
| 1156 |
+
"loss": 1.5216,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6513752358256379,
|
| 1161 |
+
"grad_norm": 0.8295606496141018,
|
| 1162 |
+
"learning_rate": 6.5479440928735695e-06,
|
| 1163 |
+
"loss": 1.5388,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.655347036044087,
|
| 1168 |
+
"grad_norm": 0.8553791665518854,
|
| 1169 |
+
"learning_rate": 6.418047850567699e-06,
|
| 1170 |
+
"loss": 1.5045,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.659318836262536,
|
| 1175 |
+
"grad_norm": 0.8249727222480653,
|
| 1176 |
+
"learning_rate": 6.2888413074448954e-06,
|
| 1177 |
+
"loss": 1.5061,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.663290636480985,
|
| 1182 |
+
"grad_norm": 0.8486443716521056,
|
| 1183 |
+
"learning_rate": 6.160349342019299e-06,
|
| 1184 |
+
"loss": 1.5202,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.667262436699434,
|
| 1189 |
+
"grad_norm": 0.8425578204636239,
|
| 1190 |
+
"learning_rate": 6.032596695214272e-06,
|
| 1191 |
+
"loss": 1.524,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.671234236917883,
|
| 1196 |
+
"grad_norm": 0.8607932133751491,
|
| 1197 |
+
"learning_rate": 5.905607965598542e-06,
|
| 1198 |
+
"loss": 1.5054,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6752060371363321,
|
| 1203 |
+
"grad_norm": 0.87671463806252,
|
| 1204 |
+
"learning_rate": 5.779407604649794e-06,
|
| 1205 |
+
"loss": 1.5256,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6791778373547811,
|
| 1210 |
+
"grad_norm": 0.8240359232861163,
|
| 1211 |
+
"learning_rate": 5.6540199120465985e-06,
|
| 1212 |
+
"loss": 1.5311,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6831496375732301,
|
| 1217 |
+
"grad_norm": 0.8880311823283904,
|
| 1218 |
+
"learning_rate": 5.529469030989511e-06,
|
| 1219 |
+
"loss": 1.5073,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6871214377916791,
|
| 1224 |
+
"grad_norm": 0.8623990682016829,
|
| 1225 |
+
"learning_rate": 5.405778943552373e-06,
|
| 1226 |
+
"loss": 1.5379,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6910932380101281,
|
| 1231 |
+
"grad_norm": 0.8262496371808098,
|
| 1232 |
+
"learning_rate": 5.282973466064567e-06,
|
| 1233 |
+
"loss": 1.5145,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6950650382285771,
|
| 1238 |
+
"grad_norm": 0.8482342550959973,
|
| 1239 |
+
"learning_rate": 5.161076244525254e-06,
|
| 1240 |
+
"loss": 1.5275,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6990368384470261,
|
| 1245 |
+
"grad_norm": 0.8328878311342953,
|
| 1246 |
+
"learning_rate": 5.040110750050356e-06,
|
| 1247 |
+
"loss": 1.528,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.7030086386654751,
|
| 1252 |
+
"grad_norm": 0.8164657341322779,
|
| 1253 |
+
"learning_rate": 4.9201002743532175e-06,
|
| 1254 |
+
"loss": 1.5092,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.7069804388839241,
|
| 1259 |
+
"grad_norm": 0.8490381133150996,
|
| 1260 |
+
"learning_rate": 4.801067925259857e-06,
|
| 1261 |
+
"loss": 1.5454,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.7109522391023732,
|
| 1266 |
+
"grad_norm": 0.8440725903075691,
|
| 1267 |
+
"learning_rate": 4.683036622259562e-06,
|
| 1268 |
+
"loss": 1.5224,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.7149240393208222,
|
| 1273 |
+
"grad_norm": 0.8366180827607501,
|
| 1274 |
+
"learning_rate": 4.566029092091798e-06,
|
| 1275 |
+
"loss": 1.5413,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.7188958395392712,
|
| 1280 |
+
"grad_norm": 0.8427766816018408,
|
| 1281 |
+
"learning_rate": 4.450067864370187e-06,
|
| 1282 |
+
"loss": 1.5173,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.7228676397577202,
|
| 1287 |
+
"grad_norm": 0.8028067037799445,
|
| 1288 |
+
"learning_rate": 4.335175267244495e-06,
|
| 1289 |
+
"loss": 1.5275,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7268394399761692,
|
| 1294 |
+
"grad_norm": 0.795885144548523,
|
| 1295 |
+
"learning_rate": 4.221373423101356e-06,
|
| 1296 |
+
"loss": 1.5166,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7308112401946182,
|
| 1301 |
+
"grad_norm": 0.8378592839978902,
|
| 1302 |
+
"learning_rate": 4.108684244304668e-06,
|
| 1303 |
+
"loss": 1.5444,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7347830404130672,
|
| 1308 |
+
"grad_norm": 0.8299609184798702,
|
| 1309 |
+
"learning_rate": 3.997129428976374e-06,
|
| 1310 |
+
"loss": 1.5373,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.7387548406315162,
|
| 1315 |
+
"grad_norm": 0.8200382060222382,
|
| 1316 |
+
"learning_rate": 3.886730456818546e-06,
|
| 1317 |
+
"loss": 1.5203,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.7427266408499652,
|
| 1322 |
+
"grad_norm": 0.8198010231353143,
|
| 1323 |
+
"learning_rate": 3.7775085849774973e-06,
|
| 1324 |
+
"loss": 1.5291,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.7466984410684142,
|
| 1329 |
+
"grad_norm": 0.8232628031861067,
|
| 1330 |
+
"learning_rate": 3.6694848439507347e-06,
|
| 1331 |
+
"loss": 1.5334,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7506702412868632,
|
| 1336 |
+
"grad_norm": 0.81406141716096,
|
| 1337 |
+
"learning_rate": 3.562680033537598e-06,
|
| 1338 |
+
"loss": 1.5091,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7546420415053123,
|
| 1343 |
+
"grad_norm": 0.8087303425914779,
|
| 1344 |
+
"learning_rate": 3.457114718834278e-06,
|
| 1345 |
+
"loss": 1.5313,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7586138417237613,
|
| 1350 |
+
"grad_norm": 0.8083510625412864,
|
| 1351 |
+
"learning_rate": 3.3528092262740284e-06,
|
| 1352 |
+
"loss": 1.5234,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7625856419422103,
|
| 1357 |
+
"grad_norm": 0.8366676453174993,
|
| 1358 |
+
"learning_rate": 3.2497836397133663e-06,
|
| 1359 |
+
"loss": 1.538,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7665574421606594,
|
| 1364 |
+
"grad_norm": 0.8393538795255264,
|
| 1365 |
+
"learning_rate": 3.148057796564944e-06,
|
| 1366 |
+
"loss": 1.5216,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.7705292423791084,
|
| 1371 |
+
"grad_norm": 0.8520167906125237,
|
| 1372 |
+
"learning_rate": 3.0476512839778894e-06,
|
| 1373 |
+
"loss": 1.5293,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7745010425975574,
|
| 1378 |
+
"grad_norm": 0.8371019471875751,
|
| 1379 |
+
"learning_rate": 2.948583435066329e-06,
|
| 1380 |
+
"loss": 1.513,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7784728428160064,
|
| 1385 |
+
"grad_norm": 0.8298639126272734,
|
| 1386 |
+
"learning_rate": 2.850873325186826e-06,
|
| 1387 |
+
"loss": 1.5268,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.7824446430344554,
|
| 1392 |
+
"grad_norm": 0.8374513555969967,
|
| 1393 |
+
"learning_rate": 2.7545397682654693e-06,
|
| 1394 |
+
"loss": 1.5097,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7864164432529044,
|
| 1399 |
+
"grad_norm": 0.8200846092055774,
|
| 1400 |
+
"learning_rate": 2.6596013131752498e-06,
|
| 1401 |
+
"loss": 1.543,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7903882434713534,
|
| 1406 |
+
"grad_norm": 0.8148429888845868,
|
| 1407 |
+
"learning_rate": 2.566076240164536e-06,
|
| 1408 |
+
"loss": 1.5522,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7943600436898024,
|
| 1413 |
+
"grad_norm": 0.8131435969188188,
|
| 1414 |
+
"learning_rate": 2.47398255733722e-06,
|
| 1415 |
+
"loss": 1.5245,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7983318439082514,
|
| 1420 |
+
"grad_norm": 0.8284789559376622,
|
| 1421 |
+
"learning_rate": 2.383337997185299e-06,
|
| 1422 |
+
"loss": 1.5427,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.8023036441267004,
|
| 1427 |
+
"grad_norm": 0.8321861504525199,
|
| 1428 |
+
"learning_rate": 2.2941600131744978e-06,
|
| 1429 |
+
"loss": 1.5072,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.8062754443451494,
|
| 1434 |
+
"grad_norm": 0.8065047141388554,
|
| 1435 |
+
"learning_rate": 2.2064657763836474e-06,
|
| 1436 |
+
"loss": 1.5179,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.8102472445635984,
|
| 1441 |
+
"grad_norm": 0.7992276385422232,
|
| 1442 |
+
"learning_rate": 2.120272172198412e-06,
|
| 1443 |
+
"loss": 1.5272,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.8142190447820474,
|
| 1448 |
+
"grad_norm": 0.8229954286323785,
|
| 1449 |
+
"learning_rate": 2.0355957970600406e-06,
|
| 1450 |
+
"loss": 1.5027,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.8181908450004964,
|
| 1455 |
+
"grad_norm": 0.8275865130237602,
|
| 1456 |
+
"learning_rate": 1.952452955269738e-06,
|
| 1457 |
+
"loss": 1.5267,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.8221626452189454,
|
| 1462 |
+
"grad_norm": 0.8081049582664315,
|
| 1463 |
+
"learning_rate": 1.870859655849304e-06,
|
| 1464 |
+
"loss": 1.5426,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.8261344454373944,
|
| 1469 |
+
"grad_norm": 0.823108756311465,
|
| 1470 |
+
"learning_rate": 1.7908316094586343e-06,
|
| 1471 |
+
"loss": 1.5388,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.8301062456558436,
|
| 1476 |
+
"grad_norm": 0.8355851750630425,
|
| 1477 |
+
"learning_rate": 1.7123842253706302e-06,
|
| 1478 |
+
"loss": 1.5014,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.8340780458742926,
|
| 1483 |
+
"grad_norm": 0.8374873929351686,
|
| 1484 |
+
"learning_rate": 1.6355326085041944e-06,
|
| 1485 |
+
"loss": 1.5008,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.8380498460927416,
|
| 1490 |
+
"grad_norm": 0.8230242479079336,
|
| 1491 |
+
"learning_rate": 1.5602915565157895e-06,
|
| 1492 |
+
"loss": 1.5249,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.8420216463111906,
|
| 1497 |
+
"grad_norm": 0.8028112832237307,
|
| 1498 |
+
"learning_rate": 1.4866755569501757e-06,
|
| 1499 |
+
"loss": 1.5134,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.8459934465296396,
|
| 1504 |
+
"grad_norm": 0.7982837438136364,
|
| 1505 |
+
"learning_rate": 1.4146987844508509e-06,
|
| 1506 |
+
"loss": 1.5127,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.8499652467480886,
|
| 1511 |
+
"grad_norm": 0.799526922604763,
|
| 1512 |
+
"learning_rate": 1.344375098030759e-06,
|
| 1513 |
+
"loss": 1.5145,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.8539370469665376,
|
| 1518 |
+
"grad_norm": 0.8028491606704994,
|
| 1519 |
+
"learning_rate": 1.2757180384037505e-06,
|
| 1520 |
+
"loss": 1.5601,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.8579088471849866,
|
| 1525 |
+
"grad_norm": 0.8062739981966481,
|
| 1526 |
+
"learning_rate": 1.2087408253773326e-06,
|
| 1527 |
+
"loss": 1.5104,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.8618806474034356,
|
| 1532 |
+
"grad_norm": 0.810426514136181,
|
| 1533 |
+
"learning_rate": 1.1434563553072332e-06,
|
| 1534 |
+
"loss": 1.5271,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.8658524476218846,
|
| 1539 |
+
"grad_norm": 0.8106217030883788,
|
| 1540 |
+
"learning_rate": 1.0798771986142186e-06,
|
| 1541 |
+
"loss": 1.5275,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.8698242478403336,
|
| 1546 |
+
"grad_norm": 0.8239831789081276,
|
| 1547 |
+
"learning_rate": 1.018015597363673e-06,
|
| 1548 |
+
"loss": 1.5435,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.8737960480587826,
|
| 1553 |
+
"grad_norm": 0.8032988356076526,
|
| 1554 |
+
"learning_rate": 9.578834629084199e-07,
|
| 1555 |
+
"loss": 1.5006,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.8777678482772316,
|
| 1560 |
+
"grad_norm": 0.8107500288685499,
|
| 1561 |
+
"learning_rate": 8.994923735952033e-07,
|
| 1562 |
+
"loss": 1.5175,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.8817396484956807,
|
| 1567 |
+
"grad_norm": 0.7947044768084438,
|
| 1568 |
+
"learning_rate": 8.428535725353016e-07,
|
| 1569 |
+
"loss": 1.5134,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.8857114487141297,
|
| 1574 |
+
"grad_norm": 0.8184180092012469,
|
| 1575 |
+
"learning_rate": 7.879779654396724e-07,
|
| 1576 |
+
"loss": 1.5308,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.8896832489325787,
|
| 1581 |
+
"grad_norm": 0.7812672670695122,
|
| 1582 |
+
"learning_rate": 7.348761185190889e-07,
|
| 1583 |
+
"loss": 1.5286,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8936550491510277,
|
| 1588 |
+
"grad_norm": 0.8031429398431277,
|
| 1589 |
+
"learning_rate": 6.835582564496257e-07,
|
| 1590 |
+
"loss": 1.5189,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8976268493694767,
|
| 1595 |
+
"grad_norm": 0.8262986271960238,
|
| 1596 |
+
"learning_rate": 6.340342604039151e-07,
|
| 1597 |
+
"loss": 1.5137,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.9015986495879257,
|
| 1602 |
+
"grad_norm": 0.8091884297906192,
|
| 1603 |
+
"learning_rate": 5.86313666148538e-07,
|
| 1604 |
+
"loss": 1.5287,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.9055704498063747,
|
| 1609 |
+
"grad_norm": 0.8083089870588875,
|
| 1610 |
+
"learning_rate": 5.40405662207939e-07,
|
| 1611 |
+
"loss": 1.5211,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.9095422500248238,
|
| 1616 |
+
"grad_norm": 0.8080847975437143,
|
| 1617 |
+
"learning_rate": 4.963190880951807e-07,
|
| 1618 |
+
"loss": 1.4942,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.9135140502432728,
|
| 1623 |
+
"grad_norm": 0.8112755205962424,
|
| 1624 |
+
"learning_rate": 4.540624326099108e-07,
|
| 1625 |
+
"loss": 1.5361,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.9174858504617218,
|
| 1630 |
+
"grad_norm": 0.8137159093665213,
|
| 1631 |
+
"learning_rate": 4.136438322038594e-07,
|
| 1632 |
+
"loss": 1.5305,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.9214576506801708,
|
| 1637 |
+
"grad_norm": 0.8099723012257717,
|
| 1638 |
+
"learning_rate": 3.7507106941418127e-07,
|
| 1639 |
+
"loss": 1.5136,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.9254294508986198,
|
| 1644 |
+
"grad_norm": 0.7874754581462157,
|
| 1645 |
+
"learning_rate": 3.3835157136493434e-07,
|
| 1646 |
+
"loss": 1.5192,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.9294012511170688,
|
| 1651 |
+
"grad_norm": 0.8097635833636108,
|
| 1652 |
+
"learning_rate": 3.0349240833699944e-07,
|
| 1653 |
+
"loss": 1.53,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.9333730513355178,
|
| 1658 |
+
"grad_norm": 0.7839597922407979,
|
| 1659 |
+
"learning_rate": 2.7050029240671573e-07,
|
| 1660 |
+
"loss": 1.5406,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.9373448515539669,
|
| 1665 |
+
"grad_norm": 0.8036152094174389,
|
| 1666 |
+
"learning_rate": 2.3938157615347455e-07,
|
| 1667 |
+
"loss": 1.5142,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.9413166517724159,
|
| 1672 |
+
"grad_norm": 0.7969437729501767,
|
| 1673 |
+
"learning_rate": 2.1014225143654145e-07,
|
| 1674 |
+
"loss": 1.5168,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.9452884519908649,
|
| 1679 |
+
"grad_norm": 0.7986549378127612,
|
| 1680 |
+
"learning_rate": 1.8278794824133417e-07,
|
| 1681 |
+
"loss": 1.5233,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.9492602522093139,
|
| 1686 |
+
"grad_norm": 0.807465267893701,
|
| 1687 |
+
"learning_rate": 1.5732393359537912e-07,
|
| 1688 |
+
"loss": 1.5193,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.9532320524277629,
|
| 1693 |
+
"grad_norm": 0.8289613150037851,
|
| 1694 |
+
"learning_rate": 1.3375511055415346e-07,
|
| 1695 |
+
"loss": 1.5306,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.9572038526462119,
|
| 1700 |
+
"grad_norm": 0.8200430426978503,
|
| 1701 |
+
"learning_rate": 1.1208601725700374e-07,
|
| 1702 |
+
"loss": 1.5185,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.9611756528646609,
|
| 1707 |
+
"grad_norm": 0.8207956989586505,
|
| 1708 |
+
"learning_rate": 9.232082605334369e-08,
|
| 1709 |
+
"loss": 1.5261,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.9651474530831099,
|
| 1714 |
+
"grad_norm": 0.8115966572361262,
|
| 1715 |
+
"learning_rate": 7.446334269926803e-08,
|
| 1716 |
+
"loss": 1.5243,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.9691192533015589,
|
| 1721 |
+
"grad_norm": 0.7861487020593475,
|
| 1722 |
+
"learning_rate": 5.851700562476304e-08,
|
| 1723 |
+
"loss": 1.5168,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.9730910535200079,
|
| 1728 |
+
"grad_norm": 0.7965492838019634,
|
| 1729 |
+
"learning_rate": 4.448488527164174e-08,
|
| 1730 |
+
"loss": 1.5238,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.9770628537384569,
|
| 1735 |
+
"grad_norm": 0.7940697296490887,
|
| 1736 |
+
"learning_rate": 3.236968350233793e-08,
|
| 1737 |
+
"loss": 1.5131,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.9810346539569059,
|
| 1742 |
+
"grad_norm": 0.8173719106280355,
|
| 1743 |
+
"learning_rate": 2.217373307966009e-08,
|
| 1744 |
+
"loss": 1.5142,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.9850064541753549,
|
| 1749 |
+
"grad_norm": 0.7895572135274385,
|
| 1750 |
+
"learning_rate": 1.389899721762844e-08,
|
| 1751 |
+
"loss": 1.5156,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.988978254393804,
|
| 1756 |
+
"grad_norm": 0.7929243131364883,
|
| 1757 |
+
"learning_rate": 7.54706920345738e-09,
|
| 1758 |
+
"loss": 1.5024,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.992950054612253,
|
| 1763 |
+
"grad_norm": 0.7966487036007772,
|
| 1764 |
+
"learning_rate": 3.119172090771949e-09,
|
| 1765 |
+
"loss": 1.5061,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.9969218548307021,
|
| 1770 |
+
"grad_norm": 0.779081898680904,
|
| 1771 |
+
"learning_rate": 6.161584641128926e-10,
|
| 1772 |
+
"loss": 1.526,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.9993049349617714,
|
| 1777 |
+
"eval_loss": 1.5326507091522217,
|
| 1778 |
+
"eval_runtime": 268.2952,
|
| 1779 |
+
"eval_samples_per_second": 99.681,
|
| 1780 |
+
"eval_steps_per_second": 4.156,
|
| 1781 |
+
"step": 1258
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9993049349617714,
|
| 1785 |
+
"step": 1258,
|
| 1786 |
+
"total_flos": 106161864966144.0,
|
| 1787 |
+
"train_loss": 1.5477431688475496,
|
| 1788 |
+
"train_runtime": 10888.6684,
|
| 1789 |
+
"train_samples_per_second": 22.196,
|
| 1790 |
+
"train_steps_per_second": 0.116
|
| 1791 |
+
}
|
| 1792 |
+
],
|
| 1793 |
+
"logging_steps": 5,
|
| 1794 |
+
"max_steps": 1258,
|
| 1795 |
+
"num_input_tokens_seen": 0,
|
| 1796 |
+
"num_train_epochs": 1,
|
| 1797 |
+
"save_steps": 500,
|
| 1798 |
+
"stateful_callbacks": {
|
| 1799 |
+
"TrainerControl": {
|
| 1800 |
+
"args": {
|
| 1801 |
+
"should_epoch_stop": false,
|
| 1802 |
+
"should_evaluate": false,
|
| 1803 |
+
"should_log": false,
|
| 1804 |
+
"should_save": false,
|
| 1805 |
+
"should_training_stop": false
|
| 1806 |
+
},
|
| 1807 |
+
"attributes": {}
|
| 1808 |
+
}
|
| 1809 |
+
},
|
| 1810 |
+
"total_flos": 106161864966144.0,
|
| 1811 |
+
"train_batch_size": 8,
|
| 1812 |
+
"trial_name": null,
|
| 1813 |
+
"trial_params": null
|
| 1814 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9abc165c32397340ae7b0df34154aa010f5d2832befe728de411fabdea7ece72
|
| 3 |
+
size 6904
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|