Commit
•
913f6f3
1
Parent(s):
bc465bc
Model save
Browse files- README.md +68 -0
- all_results.json +9 -0
- config.json +30 -0
- generation_config.json +8 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- runs/Aug27_09-54-50_GPU-4090s7/events.out.tfevents.1724745293.GPU-4090s7.3323851.0 +3 -0
- special_tokens_map.json +28 -0
- tokenizer.json +0 -0
- tokenizer_config.json +154 -0
- train_results.json +9 -0
- trainer_state.json +2262 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: HuggingFaceTB/SmolLM-360M-Instruct
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: smollm-360M-aliases-5-plus
|
| 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 |
+
# smollm-360M-aliases-5-plus
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [HuggingFaceTB/SmolLM-360M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM-360M-Instruct) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 2.2459
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.001
|
| 43 |
+
- train_batch_size: 4
|
| 44 |
+
- eval_batch_size: 4
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 2
|
| 48 |
+
- gradient_accumulation_steps: 4
|
| 49 |
+
- total_train_batch_size: 32
|
| 50 |
+
- total_eval_batch_size: 8
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 54 |
+
- num_epochs: 1
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 0.956 | 0.9997 | 1576 | 2.2459 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- Transformers 4.44.2
|
| 66 |
+
- Pytorch 2.4.0+cu121
|
| 67 |
+
- Datasets 2.21.0
|
| 68 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9996828417380272,
|
| 3 |
+
"total_flos": 38663670988800.0,
|
| 4 |
+
"train_loss": 1.4483577938854393,
|
| 5 |
+
"train_runtime": 3409.163,
|
| 6 |
+
"train_samples": 750000,
|
| 7 |
+
"train_samples_per_second": 14.797,
|
| 8 |
+
"train_steps_per_second": 0.462
|
| 9 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "HuggingFaceTB/SmolLM-360M-Instruct",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 960,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 2560,
|
| 14 |
+
"max_position_embeddings": 2048,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
+
"model_type": "llama",
|
| 17 |
+
"num_attention_heads": 15,
|
| 18 |
+
"num_hidden_layers": 32,
|
| 19 |
+
"num_key_value_heads": 5,
|
| 20 |
+
"pad_token_id": 2,
|
| 21 |
+
"pretraining_tp": 1,
|
| 22 |
+
"rms_norm_eps": 1e-05,
|
| 23 |
+
"rope_scaling": null,
|
| 24 |
+
"rope_theta": 10000.0,
|
| 25 |
+
"tie_word_embeddings": true,
|
| 26 |
+
"torch_dtype": "bfloat16",
|
| 27 |
+
"transformers_version": "4.44.2",
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"vocab_size": 49152
|
| 30 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"max_new_tokens": 40,
|
| 6 |
+
"pad_token_id": 2,
|
| 7 |
+
"transformers_version": "4.44.2"
|
| 8 |
+
}
|
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 sha256:74c45941ffa5c383e215ed07c0e441952403629fb3f16bdd5359f0703fea7c2f
|
| 3 |
+
size 723674912
|
runs/Aug27_09-54-50_GPU-4090s7/events.out.tfevents.1724745293.GPU-4090s7.3323851.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cbedf05e00632c8e5bddfd447ad8e72f111b8667f5bee993580e479ea3ffdea8
|
| 3 |
+
size 72668
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"unk_token": {
|
| 22 |
+
"content": "<|endoftext|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false
|
| 27 |
+
}
|
| 28 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<repo_name>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"4": {
|
| 37 |
+
"content": "<reponame>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"5": {
|
| 45 |
+
"content": "<file_sep>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"6": {
|
| 53 |
+
"content": "<filename>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"7": {
|
| 61 |
+
"content": "<gh_stars>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
},
|
| 68 |
+
"8": {
|
| 69 |
+
"content": "<issue_start>",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": false,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": true
|
| 75 |
+
},
|
| 76 |
+
"9": {
|
| 77 |
+
"content": "<issue_comment>",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": false,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": true
|
| 83 |
+
},
|
| 84 |
+
"10": {
|
| 85 |
+
"content": "<issue_closed>",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": false,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
"11": {
|
| 93 |
+
"content": "<jupyter_start>",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": false,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": true
|
| 99 |
+
},
|
| 100 |
+
"12": {
|
| 101 |
+
"content": "<jupyter_text>",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": false,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": true
|
| 107 |
+
},
|
| 108 |
+
"13": {
|
| 109 |
+
"content": "<jupyter_code>",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": false,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": true
|
| 115 |
+
},
|
| 116 |
+
"14": {
|
| 117 |
+
"content": "<jupyter_output>",
|
| 118 |
+
"lstrip": false,
|
| 119 |
+
"normalized": false,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": true
|
| 123 |
+
},
|
| 124 |
+
"15": {
|
| 125 |
+
"content": "<jupyter_script>",
|
| 126 |
+
"lstrip": false,
|
| 127 |
+
"normalized": false,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": true
|
| 131 |
+
},
|
| 132 |
+
"16": {
|
| 133 |
+
"content": "<empty_output>",
|
| 134 |
+
"lstrip": false,
|
| 135 |
+
"normalized": false,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": true
|
| 139 |
+
}
|
| 140 |
+
},
|
| 141 |
+
"additional_special_tokens": [
|
| 142 |
+
"<|im_start|>",
|
| 143 |
+
"<|im_end|>"
|
| 144 |
+
],
|
| 145 |
+
"bos_token": "<|im_start|>",
|
| 146 |
+
"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 %}",
|
| 147 |
+
"clean_up_tokenization_spaces": false,
|
| 148 |
+
"eos_token": "<|im_end|>",
|
| 149 |
+
"model_max_length": 2048,
|
| 150 |
+
"pad_token": "<|im_end|>",
|
| 151 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 152 |
+
"unk_token": "<|endoftext|>",
|
| 153 |
+
"vocab_size": 49152
|
| 154 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9996828417380272,
|
| 3 |
+
"total_flos": 38663670988800.0,
|
| 4 |
+
"train_loss": 1.4483577938854393,
|
| 5 |
+
"train_runtime": 3409.163,
|
| 6 |
+
"train_samples": 750000,
|
| 7 |
+
"train_samples_per_second": 14.797,
|
| 8 |
+
"train_steps_per_second": 0.462
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2262 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9996828417380272,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1576,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006343165239454488,
|
| 13 |
+
"grad_norm": 4.072216572643748,
|
| 14 |
+
"learning_rate": 6.329113924050633e-06,
|
| 15 |
+
"loss": 3.2618,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003171582619727244,
|
| 20 |
+
"grad_norm": 3.6569951400637444,
|
| 21 |
+
"learning_rate": 3.1645569620253167e-05,
|
| 22 |
+
"loss": 3.3026,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006343165239454488,
|
| 27 |
+
"grad_norm": 2.7101736291338634,
|
| 28 |
+
"learning_rate": 6.329113924050633e-05,
|
| 29 |
+
"loss": 3.1879,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009514747859181731,
|
| 34 |
+
"grad_norm": 0.8292602783930105,
|
| 35 |
+
"learning_rate": 9.49367088607595e-05,
|
| 36 |
+
"loss": 2.9431,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012686330478908976,
|
| 41 |
+
"grad_norm": 0.4014812685185947,
|
| 42 |
+
"learning_rate": 0.00012658227848101267,
|
| 43 |
+
"loss": 2.7946,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01585791309863622,
|
| 48 |
+
"grad_norm": 0.3318705203128404,
|
| 49 |
+
"learning_rate": 0.00015822784810126583,
|
| 50 |
+
"loss": 2.6345,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.019029495718363463,
|
| 55 |
+
"grad_norm": 0.4987722149702972,
|
| 56 |
+
"learning_rate": 0.000189873417721519,
|
| 57 |
+
"loss": 2.547,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.022201078338090707,
|
| 62 |
+
"grad_norm": 0.3868479554032603,
|
| 63 |
+
"learning_rate": 0.00022151898734177215,
|
| 64 |
+
"loss": 2.462,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.02537266095781795,
|
| 69 |
+
"grad_norm": 0.3440357488186946,
|
| 70 |
+
"learning_rate": 0.00025316455696202533,
|
| 71 |
+
"loss": 2.4119,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.028544243577545196,
|
| 76 |
+
"grad_norm": 0.2542091158949379,
|
| 77 |
+
"learning_rate": 0.0002848101265822785,
|
| 78 |
+
"loss": 2.3337,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03171582619727244,
|
| 83 |
+
"grad_norm": 0.5028112177822042,
|
| 84 |
+
"learning_rate": 0.00031645569620253165,
|
| 85 |
+
"loss": 2.3079,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.034887408816999685,
|
| 90 |
+
"grad_norm": 0.3647071858886548,
|
| 91 |
+
"learning_rate": 0.00034810126582278487,
|
| 92 |
+
"loss": 2.2772,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.038058991436726926,
|
| 97 |
+
"grad_norm": 0.3365279113242041,
|
| 98 |
+
"learning_rate": 0.000379746835443038,
|
| 99 |
+
"loss": 2.2643,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.041230574056454174,
|
| 104 |
+
"grad_norm": 0.3437920074360608,
|
| 105 |
+
"learning_rate": 0.0004113924050632912,
|
| 106 |
+
"loss": 2.2467,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.044402156676181415,
|
| 111 |
+
"grad_norm": 0.2180989181512006,
|
| 112 |
+
"learning_rate": 0.0004430379746835443,
|
| 113 |
+
"loss": 2.2192,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.047573739295908656,
|
| 118 |
+
"grad_norm": 0.3740832044447792,
|
| 119 |
+
"learning_rate": 0.00047468354430379745,
|
| 120 |
+
"loss": 2.2202,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0507453219156359,
|
| 125 |
+
"grad_norm": 0.4042428788012064,
|
| 126 |
+
"learning_rate": 0.0005063291139240507,
|
| 127 |
+
"loss": 2.206,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.053916904535363144,
|
| 132 |
+
"grad_norm": 0.5502810429404877,
|
| 133 |
+
"learning_rate": 0.0005379746835443038,
|
| 134 |
+
"loss": 2.1792,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05708848715509039,
|
| 139 |
+
"grad_norm": 0.9725610364599878,
|
| 140 |
+
"learning_rate": 0.000569620253164557,
|
| 141 |
+
"loss": 2.1717,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.06026006977481763,
|
| 146 |
+
"grad_norm": 0.4008236462318082,
|
| 147 |
+
"learning_rate": 0.0006012658227848101,
|
| 148 |
+
"loss": 2.1581,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06343165239454487,
|
| 153 |
+
"grad_norm": 1.1094243654374898,
|
| 154 |
+
"learning_rate": 0.0006329113924050633,
|
| 155 |
+
"loss": 2.1497,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06660323501427212,
|
| 160 |
+
"grad_norm": 0.257873202758346,
|
| 161 |
+
"learning_rate": 0.0006645569620253165,
|
| 162 |
+
"loss": 2.1357,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06977481763399937,
|
| 167 |
+
"grad_norm": 0.4530053303085577,
|
| 168 |
+
"learning_rate": 0.0006962025316455697,
|
| 169 |
+
"loss": 2.1319,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0729464002537266,
|
| 174 |
+
"grad_norm": 0.237617906265262,
|
| 175 |
+
"learning_rate": 0.0007278481012658228,
|
| 176 |
+
"loss": 2.114,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07611798287345385,
|
| 181 |
+
"grad_norm": 1.309988153168323,
|
| 182 |
+
"learning_rate": 0.000759493670886076,
|
| 183 |
+
"loss": 2.101,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0792895654931811,
|
| 188 |
+
"grad_norm": 0.33147972360135136,
|
| 189 |
+
"learning_rate": 0.0007911392405063291,
|
| 190 |
+
"loss": 2.0983,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08246114811290835,
|
| 195 |
+
"grad_norm": 0.6923953247791184,
|
| 196 |
+
"learning_rate": 0.0008227848101265824,
|
| 197 |
+
"loss": 2.0775,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08563273073263558,
|
| 202 |
+
"grad_norm": 0.6108397955993198,
|
| 203 |
+
"learning_rate": 0.0008544303797468354,
|
| 204 |
+
"loss": 2.076,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08880431335236283,
|
| 209 |
+
"grad_norm": 0.25451367249745316,
|
| 210 |
+
"learning_rate": 0.0008860759493670886,
|
| 211 |
+
"loss": 2.0613,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.09197589597209008,
|
| 216 |
+
"grad_norm": 0.5696237036165042,
|
| 217 |
+
"learning_rate": 0.0009177215189873418,
|
| 218 |
+
"loss": 2.0571,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09514747859181731,
|
| 223 |
+
"grad_norm": 0.47600183429795934,
|
| 224 |
+
"learning_rate": 0.0009493670886075949,
|
| 225 |
+
"loss": 2.0597,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09831906121154456,
|
| 230 |
+
"grad_norm": 0.49360558045014563,
|
| 231 |
+
"learning_rate": 0.0009810126582278482,
|
| 232 |
+
"loss": 2.0276,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.1014906438312718,
|
| 237 |
+
"grad_norm": 0.2619769061768294,
|
| 238 |
+
"learning_rate": 0.0009999950915251159,
|
| 239 |
+
"loss": 2.049,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.10466222645099905,
|
| 244 |
+
"grad_norm": 0.25933552218187667,
|
| 245 |
+
"learning_rate": 0.0009999398722894419,
|
| 246 |
+
"loss": 2.0304,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10783380907072629,
|
| 251 |
+
"grad_norm": 0.334897031747453,
|
| 252 |
+
"learning_rate": 0.0009998233050230736,
|
| 253 |
+
"loss": 2.0144,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.11100539169045354,
|
| 258 |
+
"grad_norm": 0.31275160877263003,
|
| 259 |
+
"learning_rate": 0.0009996454040300758,
|
| 260 |
+
"loss": 1.9773,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.11417697431018078,
|
| 265 |
+
"grad_norm": 0.5307747748580993,
|
| 266 |
+
"learning_rate": 0.0009994061911408245,
|
| 267 |
+
"loss": 1.9863,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11734855692990802,
|
| 272 |
+
"grad_norm": 0.4066591897491239,
|
| 273 |
+
"learning_rate": 0.0009991056957093295,
|
| 274 |
+
"loss": 1.9812,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.12052013954963527,
|
| 279 |
+
"grad_norm": 0.40898061678871916,
|
| 280 |
+
"learning_rate": 0.0009987439546096308,
|
| 281 |
+
"loss": 1.9983,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.12369172216936251,
|
| 286 |
+
"grad_norm": 0.6281156992876311,
|
| 287 |
+
"learning_rate": 0.0009983210122312745,
|
| 288 |
+
"loss": 1.9663,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12686330478908975,
|
| 293 |
+
"grad_norm": 0.418957878268899,
|
| 294 |
+
"learning_rate": 0.000997836920473866,
|
| 295 |
+
"loss": 1.9443,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.130034887408817,
|
| 300 |
+
"grad_norm": 0.2967784254205414,
|
| 301 |
+
"learning_rate": 0.000997291738740701,
|
| 302 |
+
"loss": 1.9496,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.13320647002854424,
|
| 307 |
+
"grad_norm": 0.32659030796363414,
|
| 308 |
+
"learning_rate": 0.0009966855339314756,
|
| 309 |
+
"loss": 1.9394,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1363780526482715,
|
| 314 |
+
"grad_norm": 0.44336971944165454,
|
| 315 |
+
"learning_rate": 0.0009960183804340781,
|
| 316 |
+
"loss": 1.9274,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.13954963526799874,
|
| 321 |
+
"grad_norm": 0.7513841493295751,
|
| 322 |
+
"learning_rate": 0.0009952903601154596,
|
| 323 |
+
"loss": 1.937,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.142721217887726,
|
| 328 |
+
"grad_norm": 0.8840919710762163,
|
| 329 |
+
"learning_rate": 0.0009945015623115897,
|
| 330 |
+
"loss": 1.9222,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1458928005074532,
|
| 335 |
+
"grad_norm": 0.3088103069995084,
|
| 336 |
+
"learning_rate": 0.000993652083816491,
|
| 337 |
+
"loss": 1.9272,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.14906438312718046,
|
| 342 |
+
"grad_norm": 0.3205509173648325,
|
| 343 |
+
"learning_rate": 0.0009927420288703658,
|
| 344 |
+
"loss": 1.9282,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1522359657469077,
|
| 349 |
+
"grad_norm": 0.5725328480987164,
|
| 350 |
+
"learning_rate": 0.0009917715091467998,
|
| 351 |
+
"loss": 1.9092,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.15540754836663495,
|
| 356 |
+
"grad_norm": 0.6881811737021382,
|
| 357 |
+
"learning_rate": 0.000990740643739063,
|
| 358 |
+
"loss": 1.9257,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.1585791309863622,
|
| 363 |
+
"grad_norm": 0.3928160747689014,
|
| 364 |
+
"learning_rate": 0.000989649559145493,
|
| 365 |
+
"loss": 1.9075,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.16175071360608945,
|
| 370 |
+
"grad_norm": 0.3457762887590973,
|
| 371 |
+
"learning_rate": 0.000988498389253972,
|
| 372 |
+
"loss": 1.8954,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1649222962258167,
|
| 377 |
+
"grad_norm": 0.6130106535523941,
|
| 378 |
+
"learning_rate": 0.0009872872753254995,
|
| 379 |
+
"loss": 1.8869,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.16809387884554391,
|
| 384 |
+
"grad_norm": 0.2006935276789736,
|
| 385 |
+
"learning_rate": 0.0009860163659768566,
|
| 386 |
+
"loss": 1.8764,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.17126546146527116,
|
| 391 |
+
"grad_norm": 0.2519600708036085,
|
| 392 |
+
"learning_rate": 0.0009846858171623687,
|
| 393 |
+
"loss": 1.8592,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.1744370440849984,
|
| 398 |
+
"grad_norm": 0.21430434382675823,
|
| 399 |
+
"learning_rate": 0.0009832957921547696,
|
| 400 |
+
"loss": 1.8588,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.17760862670472566,
|
| 405 |
+
"grad_norm": 0.7316176065198735,
|
| 406 |
+
"learning_rate": 0.000981846461525165,
|
| 407 |
+
"loss": 1.8442,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1807802093244529,
|
| 412 |
+
"grad_norm": 0.5438158046657656,
|
| 413 |
+
"learning_rate": 0.0009803380031221018,
|
| 414 |
+
"loss": 1.8681,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.18395179194418015,
|
| 419 |
+
"grad_norm": 0.22290789589006946,
|
| 420 |
+
"learning_rate": 0.000978770602049745,
|
| 421 |
+
"loss": 1.8342,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.1871233745639074,
|
| 426 |
+
"grad_norm": 0.2561355818352734,
|
| 427 |
+
"learning_rate": 0.0009771444506451621,
|
| 428 |
+
"loss": 1.8408,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.19029495718363462,
|
| 433 |
+
"grad_norm": 0.3381776052738623,
|
| 434 |
+
"learning_rate": 0.0009754597484547223,
|
| 435 |
+
"loss": 1.829,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.19346653980336187,
|
| 440 |
+
"grad_norm": 0.2267569653346989,
|
| 441 |
+
"learning_rate": 0.0009737167022096094,
|
| 442 |
+
"loss": 1.8283,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.19663812242308912,
|
| 447 |
+
"grad_norm": 0.23165580428938548,
|
| 448 |
+
"learning_rate": 0.0009719155258004541,
|
| 449 |
+
"loss": 1.8071,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.19980970504281637,
|
| 454 |
+
"grad_norm": 0.25586494282771377,
|
| 455 |
+
"learning_rate": 0.0009700564402510871,
|
| 456 |
+
"loss": 1.8145,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.2029812876625436,
|
| 461 |
+
"grad_norm": 0.2540371949506308,
|
| 462 |
+
"learning_rate": 0.0009681396736914168,
|
| 463 |
+
"loss": 1.8015,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.20615287028227086,
|
| 468 |
+
"grad_norm": 0.6388348558478815,
|
| 469 |
+
"learning_rate": 0.0009661654613294355,
|
| 470 |
+
"loss": 1.8127,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2093244529019981,
|
| 475 |
+
"grad_norm": 0.3864015903258655,
|
| 476 |
+
"learning_rate": 0.0009641340454223575,
|
| 477 |
+
"loss": 1.7935,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.21249603552172533,
|
| 482 |
+
"grad_norm": 0.2751489116810319,
|
| 483 |
+
"learning_rate": 0.0009620456752468903,
|
| 484 |
+
"loss": 1.8058,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.21566761814145258,
|
| 489 |
+
"grad_norm": 0.7422837235361598,
|
| 490 |
+
"learning_rate": 0.0009599006070686467,
|
| 491 |
+
"loss": 1.7927,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.21883920076117983,
|
| 496 |
+
"grad_norm": 0.4534291854575538,
|
| 497 |
+
"learning_rate": 0.0009576991041106973,
|
| 498 |
+
"loss": 1.7927,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.22201078338090707,
|
| 503 |
+
"grad_norm": 0.3583139746684532,
|
| 504 |
+
"learning_rate": 0.0009554414365212709,
|
| 505 |
+
"loss": 1.7883,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.22518236600063432,
|
| 510 |
+
"grad_norm": 0.20634161577455162,
|
| 511 |
+
"learning_rate": 0.0009531278813406046,
|
| 512 |
+
"loss": 1.7637,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.22835394862036157,
|
| 517 |
+
"grad_norm": 0.5462716024749192,
|
| 518 |
+
"learning_rate": 0.000950758722466947,
|
| 519 |
+
"loss": 1.7823,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.23152553124008882,
|
| 524 |
+
"grad_norm": 0.20847302955466993,
|
| 525 |
+
"learning_rate": 0.0009483342506217214,
|
| 526 |
+
"loss": 1.7736,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.23469711385981604,
|
| 531 |
+
"grad_norm": 0.21809684764751344,
|
| 532 |
+
"learning_rate": 0.0009458547633138515,
|
| 533 |
+
"loss": 1.7636,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.23786869647954328,
|
| 538 |
+
"grad_norm": 0.19220401784144317,
|
| 539 |
+
"learning_rate": 0.0009433205648032528,
|
| 540 |
+
"loss": 1.7509,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.24104027909927053,
|
| 545 |
+
"grad_norm": 0.273271874095809,
|
| 546 |
+
"learning_rate": 0.0009407319660634979,
|
| 547 |
+
"loss": 1.7488,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.24421186171899778,
|
| 552 |
+
"grad_norm": 0.31458786826276625,
|
| 553 |
+
"learning_rate": 0.0009380892847436555,
|
| 554 |
+
"loss": 1.7342,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.24738344433872503,
|
| 559 |
+
"grad_norm": 0.19789392284188642,
|
| 560 |
+
"learning_rate": 0.0009353928451293121,
|
| 561 |
+
"loss": 1.743,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2505550269584523,
|
| 566 |
+
"grad_norm": 0.24499888411428472,
|
| 567 |
+
"learning_rate": 0.0009326429781027789,
|
| 568 |
+
"loss": 1.7193,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2537266095781795,
|
| 573 |
+
"grad_norm": 0.33173879702411524,
|
| 574 |
+
"learning_rate": 0.0009298400211024877,
|
| 575 |
+
"loss": 1.729,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.25689819219790677,
|
| 580 |
+
"grad_norm": 0.34879621136110506,
|
| 581 |
+
"learning_rate": 0.0009269843180815853,
|
| 582 |
+
"loss": 1.7241,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.260069774817634,
|
| 587 |
+
"grad_norm": 0.20572899222991778,
|
| 588 |
+
"learning_rate": 0.0009240762194657253,
|
| 589 |
+
"loss": 1.7229,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.26324135743736127,
|
| 594 |
+
"grad_norm": 0.21561079654661294,
|
| 595 |
+
"learning_rate": 0.0009211160821100679,
|
| 596 |
+
"loss": 1.7155,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2664129400570885,
|
| 601 |
+
"grad_norm": 0.443393553505543,
|
| 602 |
+
"learning_rate": 0.0009181042692554893,
|
| 603 |
+
"loss": 1.7111,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2695845226768157,
|
| 608 |
+
"grad_norm": 0.2266683853424827,
|
| 609 |
+
"learning_rate": 0.0009150411504840086,
|
| 610 |
+
"loss": 1.7009,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.272756105296543,
|
| 615 |
+
"grad_norm": 0.3582735361142464,
|
| 616 |
+
"learning_rate": 0.000911927101673436,
|
| 617 |
+
"loss": 1.7016,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2759276879162702,
|
| 622 |
+
"grad_norm": 0.43342116834945776,
|
| 623 |
+
"learning_rate": 0.0009087625049512488,
|
| 624 |
+
"loss": 1.7037,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2790992705359975,
|
| 629 |
+
"grad_norm": 0.3295875571024751,
|
| 630 |
+
"learning_rate": 0.0009055477486476991,
|
| 631 |
+
"loss": 1.682,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2822708531557247,
|
| 636 |
+
"grad_norm": 0.1891978276034803,
|
| 637 |
+
"learning_rate": 0.0009022832272481627,
|
| 638 |
+
"loss": 1.6899,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.285442435775452,
|
| 643 |
+
"grad_norm": 0.26615608448970285,
|
| 644 |
+
"learning_rate": 0.000898969341344731,
|
| 645 |
+
"loss": 1.6909,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2886140183951792,
|
| 650 |
+
"grad_norm": 0.26554406802462666,
|
| 651 |
+
"learning_rate": 0.0008956064975870544,
|
| 652 |
+
"loss": 1.6764,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2917856010149064,
|
| 657 |
+
"grad_norm": 0.20008546513645153,
|
| 658 |
+
"learning_rate": 0.0008921951086324411,
|
| 659 |
+
"loss": 1.6571,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2949571836346337,
|
| 664 |
+
"grad_norm": 0.25575390463894654,
|
| 665 |
+
"learning_rate": 0.0008887355930952202,
|
| 666 |
+
"loss": 1.6636,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2981287662543609,
|
| 671 |
+
"grad_norm": 0.3501161922386378,
|
| 672 |
+
"learning_rate": 0.0008852283754953732,
|
| 673 |
+
"loss": 1.657,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3013003488740882,
|
| 678 |
+
"grad_norm": 0.20707308621635875,
|
| 679 |
+
"learning_rate": 0.0008816738862064412,
|
| 680 |
+
"loss": 1.6659,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3044719314938154,
|
| 685 |
+
"grad_norm": 0.2572060719794171,
|
| 686 |
+
"learning_rate": 0.0008780725614027123,
|
| 687 |
+
"loss": 1.6521,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3076435141135427,
|
| 692 |
+
"grad_norm": 0.2773641851176988,
|
| 693 |
+
"learning_rate": 0.000874424843005699,
|
| 694 |
+
"loss": 1.6545,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3108150967332699,
|
| 699 |
+
"grad_norm": 0.5151669199508683,
|
| 700 |
+
"learning_rate": 0.0008707311786299099,
|
| 701 |
+
"loss": 1.6512,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3139866793529971,
|
| 706 |
+
"grad_norm": 0.35976330322294225,
|
| 707 |
+
"learning_rate": 0.0008669920215279222,
|
| 708 |
+
"loss": 1.6489,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3171582619727244,
|
| 713 |
+
"grad_norm": 0.18626964833018503,
|
| 714 |
+
"learning_rate": 0.0008632078305347623,
|
| 715 |
+
"loss": 1.6292,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3203298445924516,
|
| 720 |
+
"grad_norm": 0.26834931489718644,
|
| 721 |
+
"learning_rate": 0.0008593790700116029,
|
| 722 |
+
"loss": 1.6244,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3235014272121789,
|
| 727 |
+
"grad_norm": 0.24263398553664595,
|
| 728 |
+
"learning_rate": 0.0008555062097887796,
|
| 729 |
+
"loss": 1.6173,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3266730098319061,
|
| 734 |
+
"grad_norm": 0.20406088273168801,
|
| 735 |
+
"learning_rate": 0.0008515897251081384,
|
| 736 |
+
"loss": 1.6273,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3298445924516334,
|
| 741 |
+
"grad_norm": 0.18611145873310309,
|
| 742 |
+
"learning_rate": 0.0008476300965647186,
|
| 743 |
+
"loss": 1.5954,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3330161750713606,
|
| 748 |
+
"grad_norm": 0.2455194958653317,
|
| 749 |
+
"learning_rate": 0.0008436278100477775,
|
| 750 |
+
"loss": 1.6284,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.33618775769108783,
|
| 755 |
+
"grad_norm": 0.3089123130431731,
|
| 756 |
+
"learning_rate": 0.0008395833566811676,
|
| 757 |
+
"loss": 1.6043,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3393593403108151,
|
| 762 |
+
"grad_norm": 0.278226103442385,
|
| 763 |
+
"learning_rate": 0.0008354972327630705,
|
| 764 |
+
"loss": 1.5991,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3425309229305423,
|
| 769 |
+
"grad_norm": 0.40283564646452896,
|
| 770 |
+
"learning_rate": 0.000831369939705094,
|
| 771 |
+
"loss": 1.5942,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3457025055502696,
|
| 776 |
+
"grad_norm": 0.667879925495927,
|
| 777 |
+
"learning_rate": 0.0008272019839707461,
|
| 778 |
+
"loss": 1.5968,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3488740881699968,
|
| 783 |
+
"grad_norm": 0.34787949832129605,
|
| 784 |
+
"learning_rate": 0.0008229938770132843,
|
| 785 |
+
"loss": 1.5815,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3520456707897241,
|
| 790 |
+
"grad_norm": 0.2584482438633063,
|
| 791 |
+
"learning_rate": 0.0008187461352129555,
|
| 792 |
+
"loss": 1.5884,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.3552172534094513,
|
| 797 |
+
"grad_norm": 0.22742176340374293,
|
| 798 |
+
"learning_rate": 0.0008144592798136309,
|
| 799 |
+
"loss": 1.5919,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.35838883602917854,
|
| 804 |
+
"grad_norm": 0.42255520990843093,
|
| 805 |
+
"learning_rate": 0.0008101338368588436,
|
| 806 |
+
"loss": 1.5913,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3615604186489058,
|
| 811 |
+
"grad_norm": 0.3824113474293145,
|
| 812 |
+
"learning_rate": 0.0008057703371272366,
|
| 813 |
+
"loss": 1.5611,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.36473200126863303,
|
| 818 |
+
"grad_norm": 0.2091017723989841,
|
| 819 |
+
"learning_rate": 0.0008013693160674316,
|
| 820 |
+
"loss": 1.5626,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3679035838883603,
|
| 825 |
+
"grad_norm": 0.23814734243563393,
|
| 826 |
+
"learning_rate": 0.0007969313137323229,
|
| 827 |
+
"loss": 1.5656,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.37107516650808753,
|
| 832 |
+
"grad_norm": 0.2597004458168679,
|
| 833 |
+
"learning_rate": 0.0007924568747128076,
|
| 834 |
+
"loss": 1.5624,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3742467491278148,
|
| 839 |
+
"grad_norm": 0.2949069544402481,
|
| 840 |
+
"learning_rate": 0.0007879465480709576,
|
| 841 |
+
"loss": 1.5516,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.377418331747542,
|
| 846 |
+
"grad_norm": 0.21263382790898516,
|
| 847 |
+
"learning_rate": 0.0007834008872726453,
|
| 848 |
+
"loss": 1.5409,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.38058991436726924,
|
| 853 |
+
"grad_norm": 0.27681275720229476,
|
| 854 |
+
"learning_rate": 0.0007788204501196254,
|
| 855 |
+
"loss": 1.5507,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3837614969869965,
|
| 860 |
+
"grad_norm": 0.5196324383707882,
|
| 861 |
+
"learning_rate": 0.000774205798681088,
|
| 862 |
+
"loss": 1.5435,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.38693307960672374,
|
| 867 |
+
"grad_norm": 0.3397151418636398,
|
| 868 |
+
"learning_rate": 0.000769557499224686,
|
| 869 |
+
"loss": 1.5292,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.390104662226451,
|
| 874 |
+
"grad_norm": 0.21757261564984298,
|
| 875 |
+
"learning_rate": 0.0007648761221470481,
|
| 876 |
+
"loss": 1.5342,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.39327624484617824,
|
| 881 |
+
"grad_norm": 0.23799713493080946,
|
| 882 |
+
"learning_rate": 0.000760162241903785,
|
| 883 |
+
"loss": 1.5314,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.3964478274659055,
|
| 888 |
+
"grad_norm": 0.20955913102047505,
|
| 889 |
+
"learning_rate": 0.0007554164369389975,
|
| 890 |
+
"loss": 1.5149,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.39961941008563273,
|
| 895 |
+
"grad_norm": 0.19465193626198848,
|
| 896 |
+
"learning_rate": 0.0007506392896142951,
|
| 897 |
+
"loss": 1.514,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.40279099270535995,
|
| 902 |
+
"grad_norm": 0.37370015455345407,
|
| 903 |
+
"learning_rate": 0.0007458313861373336,
|
| 904 |
+
"loss": 1.5138,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4059625753250872,
|
| 909 |
+
"grad_norm": 0.2112845859224254,
|
| 910 |
+
"learning_rate": 0.0007409933164898818,
|
| 911 |
+
"loss": 1.5024,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.40913415794481445,
|
| 916 |
+
"grad_norm": 0.24626397881644146,
|
| 917 |
+
"learning_rate": 0.0007361256743554241,
|
| 918 |
+
"loss": 1.519,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4123057405645417,
|
| 923 |
+
"grad_norm": 0.3216374157044185,
|
| 924 |
+
"learning_rate": 0.0007312290570463083,
|
| 925 |
+
"loss": 1.5039,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.41547732318426894,
|
| 930 |
+
"grad_norm": 0.22302629969432056,
|
| 931 |
+
"learning_rate": 0.0007263040654304502,
|
| 932 |
+
"loss": 1.494,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.4186489058039962,
|
| 937 |
+
"grad_norm": 0.2675317557830398,
|
| 938 |
+
"learning_rate": 0.0007213513038575998,
|
| 939 |
+
"loss": 1.4884,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.42182048842372344,
|
| 944 |
+
"grad_norm": 0.2905992631967741,
|
| 945 |
+
"learning_rate": 0.0007163713800851811,
|
| 946 |
+
"loss": 1.4851,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.42499207104345066,
|
| 951 |
+
"grad_norm": 0.20033257450058217,
|
| 952 |
+
"learning_rate": 0.0007113649052037139,
|
| 953 |
+
"loss": 1.475,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.42816365366317793,
|
| 958 |
+
"grad_norm": 0.24204591478150614,
|
| 959 |
+
"learning_rate": 0.0007063324935618264,
|
| 960 |
+
"loss": 1.4854,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.43133523628290515,
|
| 965 |
+
"grad_norm": 0.2121430223132248,
|
| 966 |
+
"learning_rate": 0.0007012747626908679,
|
| 967 |
+
"loss": 1.4867,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.43450681890263243,
|
| 972 |
+
"grad_norm": 0.22539040426730952,
|
| 973 |
+
"learning_rate": 0.0006961923332291309,
|
| 974 |
+
"loss": 1.467,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.43767840152235965,
|
| 979 |
+
"grad_norm": 0.22695514383581045,
|
| 980 |
+
"learning_rate": 0.0006910858288456921,
|
| 981 |
+
"loss": 1.4657,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4408499841420869,
|
| 986 |
+
"grad_norm": 0.22184474318285244,
|
| 987 |
+
"learning_rate": 0.0006859558761638819,
|
| 988 |
+
"loss": 1.4423,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.44402156676181415,
|
| 993 |
+
"grad_norm": 0.29575466467956124,
|
| 994 |
+
"learning_rate": 0.0006808031046843901,
|
| 995 |
+
"loss": 1.4485,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.44719314938154137,
|
| 1000 |
+
"grad_norm": 0.21488007270980458,
|
| 1001 |
+
"learning_rate": 0.0006756281467080205,
|
| 1002 |
+
"loss": 1.4508,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.45036473200126864,
|
| 1007 |
+
"grad_norm": 0.38667469007287536,
|
| 1008 |
+
"learning_rate": 0.0006704316372580989,
|
| 1009 |
+
"loss": 1.4459,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.45353631462099586,
|
| 1014 |
+
"grad_norm": 0.5234661249173684,
|
| 1015 |
+
"learning_rate": 0.0006652142140025517,
|
| 1016 |
+
"loss": 1.435,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.45670789724072314,
|
| 1021 |
+
"grad_norm": 0.37462414488518325,
|
| 1022 |
+
"learning_rate": 0.0006599765171756538,
|
| 1023 |
+
"loss": 1.4379,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.45987947986045036,
|
| 1028 |
+
"grad_norm": 0.3040640640559466,
|
| 1029 |
+
"learning_rate": 0.0006547191894994679,
|
| 1030 |
+
"loss": 1.4341,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.46305106248017763,
|
| 1035 |
+
"grad_norm": 0.28687145037107376,
|
| 1036 |
+
"learning_rate": 0.0006494428761049736,
|
| 1037 |
+
"loss": 1.4297,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.46622264509990485,
|
| 1042 |
+
"grad_norm": 0.20728566940658133,
|
| 1043 |
+
"learning_rate": 0.0006441482244529037,
|
| 1044 |
+
"loss": 1.4124,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.4693942277196321,
|
| 1049 |
+
"grad_norm": 0.21956352378213645,
|
| 1050 |
+
"learning_rate": 0.0006388358842542938,
|
| 1051 |
+
"loss": 1.4162,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.47256581033935935,
|
| 1056 |
+
"grad_norm": 0.20961137895482168,
|
| 1057 |
+
"learning_rate": 0.0006335065073907551,
|
| 1058 |
+
"loss": 1.4055,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.47573739295908657,
|
| 1063 |
+
"grad_norm": 0.21979161995613117,
|
| 1064 |
+
"learning_rate": 0.0006281607478344823,
|
| 1065 |
+
"loss": 1.4112,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.47890897557881384,
|
| 1070 |
+
"grad_norm": 0.2707420648256881,
|
| 1071 |
+
"learning_rate": 0.0006227992615680033,
|
| 1072 |
+
"loss": 1.4127,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.48208055819854106,
|
| 1077 |
+
"grad_norm": 0.2829526420993808,
|
| 1078 |
+
"learning_rate": 0.000617422706503684,
|
| 1079 |
+
"loss": 1.3905,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.48525214081826834,
|
| 1084 |
+
"grad_norm": 0.2988909342739172,
|
| 1085 |
+
"learning_rate": 0.0006120317424029943,
|
| 1086 |
+
"loss": 1.3941,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.48842372343799556,
|
| 1091 |
+
"grad_norm": 0.2787477024270155,
|
| 1092 |
+
"learning_rate": 0.0006066270307955492,
|
| 1093 |
+
"loss": 1.404,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4915953060577228,
|
| 1098 |
+
"grad_norm": 0.22142539860110755,
|
| 1099 |
+
"learning_rate": 0.000601209234897931,
|
| 1100 |
+
"loss": 1.3886,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.49476688867745006,
|
| 1105 |
+
"grad_norm": 0.2777507592841434,
|
| 1106 |
+
"learning_rate": 0.0005957790195323064,
|
| 1107 |
+
"loss": 1.3896,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4979384712971773,
|
| 1112 |
+
"grad_norm": 0.2552429702914411,
|
| 1113 |
+
"learning_rate": 0.0005903370510448447,
|
| 1114 |
+
"loss": 1.3779,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5011100539169046,
|
| 1119 |
+
"grad_norm": 0.268362490404369,
|
| 1120 |
+
"learning_rate": 0.0005848839972239511,
|
| 1121 |
+
"loss": 1.3732,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5042816365366318,
|
| 1126 |
+
"grad_norm": 0.24326700144489616,
|
| 1127 |
+
"learning_rate": 0.0005794205272183205,
|
| 1128 |
+
"loss": 1.3748,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.507453219156359,
|
| 1133 |
+
"grad_norm": 0.36345823855975784,
|
| 1134 |
+
"learning_rate": 0.0005739473114548266,
|
| 1135 |
+
"loss": 1.3755,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5106248017760863,
|
| 1140 |
+
"grad_norm": 0.4843512449452226,
|
| 1141 |
+
"learning_rate": 0.000568465021556253,
|
| 1142 |
+
"loss": 1.3638,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5137963843958135,
|
| 1147 |
+
"grad_norm": 0.28627979209509896,
|
| 1148 |
+
"learning_rate": 0.0005629743302588779,
|
| 1149 |
+
"loss": 1.3514,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5169679670155407,
|
| 1154 |
+
"grad_norm": 0.25100043970170133,
|
| 1155 |
+
"learning_rate": 0.0005574759113299217,
|
| 1156 |
+
"loss": 1.341,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.520139549635268,
|
| 1161 |
+
"grad_norm": 0.294501664896367,
|
| 1162 |
+
"learning_rate": 0.0005519704394848692,
|
| 1163 |
+
"loss": 1.3323,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5233111322549953,
|
| 1168 |
+
"grad_norm": 0.2382026692611784,
|
| 1169 |
+
"learning_rate": 0.0005464585903046744,
|
| 1170 |
+
"loss": 1.3483,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5264827148747225,
|
| 1175 |
+
"grad_norm": 0.3233277006590301,
|
| 1176 |
+
"learning_rate": 0.0005409410401528587,
|
| 1177 |
+
"loss": 1.3275,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5296542974944497,
|
| 1182 |
+
"grad_norm": 0.27343633383605254,
|
| 1183 |
+
"learning_rate": 0.0005354184660925148,
|
| 1184 |
+
"loss": 1.3379,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.532825880114177,
|
| 1189 |
+
"grad_norm": 0.23005655087591242,
|
| 1190 |
+
"learning_rate": 0.0005298915458032233,
|
| 1191 |
+
"loss": 1.3213,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5359974627339043,
|
| 1196 |
+
"grad_norm": 0.22247227894179622,
|
| 1197 |
+
"learning_rate": 0.0005243609574978941,
|
| 1198 |
+
"loss": 1.3295,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5391690453536314,
|
| 1203 |
+
"grad_norm": 0.30014284045451645,
|
| 1204 |
+
"learning_rate": 0.0005188273798395424,
|
| 1205 |
+
"loss": 1.3214,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5423406279733587,
|
| 1210 |
+
"grad_norm": 0.3132385853038301,
|
| 1211 |
+
"learning_rate": 0.0005132914918580093,
|
| 1212 |
+
"loss": 1.3172,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.545512210593086,
|
| 1217 |
+
"grad_norm": 0.33728113255378034,
|
| 1218 |
+
"learning_rate": 0.0005077539728666374,
|
| 1219 |
+
"loss": 1.3218,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5486837932128132,
|
| 1224 |
+
"grad_norm": 0.25874007616270794,
|
| 1225 |
+
"learning_rate": 0.0005022155023789121,
|
| 1226 |
+
"loss": 1.2957,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5518553758325404,
|
| 1231 |
+
"grad_norm": 0.24203527103405384,
|
| 1232 |
+
"learning_rate": 0.0004966767600250775,
|
| 1233 |
+
"loss": 1.3035,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5550269584522677,
|
| 1238 |
+
"grad_norm": 0.21381303917505012,
|
| 1239 |
+
"learning_rate": 0.0004911384254687388,
|
| 1240 |
+
"loss": 1.2995,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.558198541071995,
|
| 1245 |
+
"grad_norm": 0.24304896972645837,
|
| 1246 |
+
"learning_rate": 0.00048560117832345984,
|
| 1247 |
+
"loss": 1.2824,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5613701236917221,
|
| 1252 |
+
"grad_norm": 0.3080366389357483,
|
| 1253 |
+
"learning_rate": 0.0004800656980693674,
|
| 1254 |
+
"loss": 1.2898,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5645417063114494,
|
| 1259 |
+
"grad_norm": 0.262557897437375,
|
| 1260 |
+
"learning_rate": 0.00047453266396977174,
|
| 1261 |
+
"loss": 1.2779,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5677132889311767,
|
| 1266 |
+
"grad_norm": 0.31083455423034645,
|
| 1267 |
+
"learning_rate": 0.00046900275498781347,
|
| 1268 |
+
"loss": 1.2806,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.570884871550904,
|
| 1273 |
+
"grad_norm": 0.21597926838814396,
|
| 1274 |
+
"learning_rate": 0.00046347664970314723,
|
| 1275 |
+
"loss": 1.274,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5740564541706311,
|
| 1280 |
+
"grad_norm": 0.22596235970597578,
|
| 1281 |
+
"learning_rate": 0.0004579550262286731,
|
| 1282 |
+
"loss": 1.2666,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5772280367903584,
|
| 1287 |
+
"grad_norm": 0.22827094422158484,
|
| 1288 |
+
"learning_rate": 0.0004524385621273246,
|
| 1289 |
+
"loss": 1.2583,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5803996194100857,
|
| 1294 |
+
"grad_norm": 0.24853325436526866,
|
| 1295 |
+
"learning_rate": 0.00044692793432892387,
|
| 1296 |
+
"loss": 1.2693,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5835712020298128,
|
| 1301 |
+
"grad_norm": 0.2765479869012326,
|
| 1302 |
+
"learning_rate": 0.00044142381904711624,
|
| 1303 |
+
"loss": 1.26,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5867427846495401,
|
| 1308 |
+
"grad_norm": 0.27285996236330706,
|
| 1309 |
+
"learning_rate": 0.00043592689169639034,
|
| 1310 |
+
"loss": 1.246,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5899143672692674,
|
| 1315 |
+
"grad_norm": 0.28781941328826144,
|
| 1316 |
+
"learning_rate": 0.0004304378268091982,
|
| 1317 |
+
"loss": 1.249,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5930859498889947,
|
| 1322 |
+
"grad_norm": 0.240504157977766,
|
| 1323 |
+
"learning_rate": 0.0004249572979531822,
|
| 1324 |
+
"loss": 1.2534,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5962575325087218,
|
| 1329 |
+
"grad_norm": 0.341483100362183,
|
| 1330 |
+
"learning_rate": 0.0004194859776485216,
|
| 1331 |
+
"loss": 1.2376,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5994291151284491,
|
| 1336 |
+
"grad_norm": 0.27130765824409686,
|
| 1337 |
+
"learning_rate": 0.0004140245372854065,
|
| 1338 |
+
"loss": 1.2426,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6026006977481764,
|
| 1343 |
+
"grad_norm": 0.28496801994375115,
|
| 1344 |
+
"learning_rate": 0.0004085736470416516,
|
| 1345 |
+
"loss": 1.2347,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6057722803679035,
|
| 1350 |
+
"grad_norm": 0.33820479660283104,
|
| 1351 |
+
"learning_rate": 0.00040313397580045765,
|
| 1352 |
+
"loss": 1.2397,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6089438629876308,
|
| 1357 |
+
"grad_norm": 0.2537502852561033,
|
| 1358 |
+
"learning_rate": 0.0003977061910683325,
|
| 1359 |
+
"loss": 1.2319,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6121154456073581,
|
| 1364 |
+
"grad_norm": 0.2543562572422921,
|
| 1365 |
+
"learning_rate": 0.0003922909588931808,
|
| 1366 |
+
"loss": 1.2221,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6152870282270854,
|
| 1371 |
+
"grad_norm": 0.28194628415561285,
|
| 1372 |
+
"learning_rate": 0.0003868889437825724,
|
| 1373 |
+
"loss": 1.2213,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6184586108468125,
|
| 1378 |
+
"grad_norm": 0.26751445743912233,
|
| 1379 |
+
"learning_rate": 0.0003815008086222007,
|
| 1380 |
+
"loss": 1.211,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.6216301934665398,
|
| 1385 |
+
"grad_norm": 0.22966413613029274,
|
| 1386 |
+
"learning_rate": 0.0003761272145945388,
|
| 1387 |
+
"loss": 1.2058,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.6248017760862671,
|
| 1392 |
+
"grad_norm": 0.24668142278345986,
|
| 1393 |
+
"learning_rate": 0.0003707688210977055,
|
| 1394 |
+
"loss": 1.2223,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.6279733587059942,
|
| 1399 |
+
"grad_norm": 0.23811743937781157,
|
| 1400 |
+
"learning_rate": 0.00036542628566455025,
|
| 1401 |
+
"loss": 1.2024,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6311449413257215,
|
| 1406 |
+
"grad_norm": 0.2901121774163334,
|
| 1407 |
+
"learning_rate": 0.0003601002638819665,
|
| 1408 |
+
"loss": 1.2036,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6343165239454488,
|
| 1413 |
+
"grad_norm": 0.2600410825499236,
|
| 1414 |
+
"learning_rate": 0.0003547914093104439,
|
| 1415 |
+
"loss": 1.2012,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6374881065651761,
|
| 1420 |
+
"grad_norm": 0.352563938838776,
|
| 1421 |
+
"learning_rate": 0.0003495003734038697,
|
| 1422 |
+
"loss": 1.1751,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6406596891849032,
|
| 1427 |
+
"grad_norm": 0.26125000772161344,
|
| 1428 |
+
"learning_rate": 0.00034422780542958827,
|
| 1429 |
+
"loss": 1.1919,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6438312718046305,
|
| 1434 |
+
"grad_norm": 0.2640437019043301,
|
| 1435 |
+
"learning_rate": 0.00033897435238872874,
|
| 1436 |
+
"loss": 1.1781,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6470028544243578,
|
| 1441 |
+
"grad_norm": 0.2782272386225361,
|
| 1442 |
+
"learning_rate": 0.00033374065893681127,
|
| 1443 |
+
"loss": 1.1821,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.650174437044085,
|
| 1448 |
+
"grad_norm": 0.24555576527657738,
|
| 1449 |
+
"learning_rate": 0.0003285273673046409,
|
| 1450 |
+
"loss": 1.1721,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6533460196638122,
|
| 1455 |
+
"grad_norm": 0.40556770599075914,
|
| 1456 |
+
"learning_rate": 0.00032333511721949817,
|
| 1457 |
+
"loss": 1.1679,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6565176022835395,
|
| 1462 |
+
"grad_norm": 0.25906084663363754,
|
| 1463 |
+
"learning_rate": 0.00031816454582663856,
|
| 1464 |
+
"loss": 1.1567,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6596891849032668,
|
| 1469 |
+
"grad_norm": 0.27183164743159954,
|
| 1470 |
+
"learning_rate": 0.0003130162876111074,
|
| 1471 |
+
"loss": 1.1596,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6628607675229939,
|
| 1476 |
+
"grad_norm": 0.24394077297020256,
|
| 1477 |
+
"learning_rate": 0.0003078909743198817,
|
| 1478 |
+
"loss": 1.1487,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6660323501427212,
|
| 1483 |
+
"grad_norm": 0.23339037702881532,
|
| 1484 |
+
"learning_rate": 0.000302789234884348,
|
| 1485 |
+
"loss": 1.1636,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6692039327624485,
|
| 1490 |
+
"grad_norm": 0.2651227300122355,
|
| 1491 |
+
"learning_rate": 0.00029771169534312583,
|
| 1492 |
+
"loss": 1.1475,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6723755153821757,
|
| 1497 |
+
"grad_norm": 0.23719809453094406,
|
| 1498 |
+
"learning_rate": 0.000292658978765246,
|
| 1499 |
+
"loss": 1.1496,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6755470980019029,
|
| 1504 |
+
"grad_norm": 0.31466276172538943,
|
| 1505 |
+
"learning_rate": 0.000287631705173693,
|
| 1506 |
+
"loss": 1.1404,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6787186806216302,
|
| 1511 |
+
"grad_norm": 0.2657362830496313,
|
| 1512 |
+
"learning_rate": 0.00028263049146932153,
|
| 1513 |
+
"loss": 1.156,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6818902632413575,
|
| 1518 |
+
"grad_norm": 0.2353420135393821,
|
| 1519 |
+
"learning_rate": 0.00027765595135515673,
|
| 1520 |
+
"loss": 1.1382,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6850618458610847,
|
| 1525 |
+
"grad_norm": 0.29180017450918116,
|
| 1526 |
+
"learning_rate": 0.00027270869526108506,
|
| 1527 |
+
"loss": 1.1403,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6882334284808119,
|
| 1532 |
+
"grad_norm": 0.28381426741820764,
|
| 1533 |
+
"learning_rate": 0.000267789330268949,
|
| 1534 |
+
"loss": 1.1351,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6914050111005392,
|
| 1539 |
+
"grad_norm": 0.2368326399732858,
|
| 1540 |
+
"learning_rate": 0.00026289846003805075,
|
| 1541 |
+
"loss": 1.1264,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.6945765937202664,
|
| 1546 |
+
"grad_norm": 0.24260754741892487,
|
| 1547 |
+
"learning_rate": 0.0002580366847310774,
|
| 1548 |
+
"loss": 1.1318,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6977481763399936,
|
| 1553 |
+
"grad_norm": 0.33032504483698477,
|
| 1554 |
+
"learning_rate": 0.0002532046009404537,
|
| 1555 |
+
"loss": 1.123,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7009197589597209,
|
| 1560 |
+
"grad_norm": 0.2626626593890248,
|
| 1561 |
+
"learning_rate": 0.00024840280161513446,
|
| 1562 |
+
"loss": 1.1147,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7040913415794482,
|
| 1567 |
+
"grad_norm": 0.24734490639888912,
|
| 1568 |
+
"learning_rate": 0.0002436318759878432,
|
| 1569 |
+
"loss": 1.1141,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7072629241991754,
|
| 1574 |
+
"grad_norm": 0.25777344330608626,
|
| 1575 |
+
"learning_rate": 0.00023889240950276602,
|
| 1576 |
+
"loss": 1.1069,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.7104345068189026,
|
| 1581 |
+
"grad_norm": 0.24965316567346824,
|
| 1582 |
+
"learning_rate": 0.00023418498374371268,
|
| 1583 |
+
"loss": 1.0961,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.7136060894386299,
|
| 1588 |
+
"grad_norm": 0.2588175173420704,
|
| 1589 |
+
"learning_rate": 0.0002295101763627483,
|
| 1590 |
+
"loss": 1.1062,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.7167776720583571,
|
| 1595 |
+
"grad_norm": 0.2617691894820057,
|
| 1596 |
+
"learning_rate": 0.00022486856100931146,
|
| 1597 |
+
"loss": 1.0949,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.7199492546780843,
|
| 1602 |
+
"grad_norm": 0.24640261475787326,
|
| 1603 |
+
"learning_rate": 0.00022026070725981867,
|
| 1604 |
+
"loss": 1.1024,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.7231208372978116,
|
| 1609 |
+
"grad_norm": 0.25512789636052857,
|
| 1610 |
+
"learning_rate": 0.0002156871805477732,
|
| 1611 |
+
"loss": 1.0981,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7262924199175389,
|
| 1616 |
+
"grad_norm": 0.2380744076277497,
|
| 1617 |
+
"learning_rate": 0.00021114854209437889,
|
| 1618 |
+
"loss": 1.0803,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.7294640025372661,
|
| 1623 |
+
"grad_norm": 0.26280691701304987,
|
| 1624 |
+
"learning_rate": 0.00020664534883967311,
|
| 1625 |
+
"loss": 1.0851,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7326355851569933,
|
| 1630 |
+
"grad_norm": 0.2565430586115982,
|
| 1631 |
+
"learning_rate": 0.00020217815337418427,
|
| 1632 |
+
"loss": 1.076,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7358071677767206,
|
| 1637 |
+
"grad_norm": 0.2533752282987412,
|
| 1638 |
+
"learning_rate": 0.00019774750387112174,
|
| 1639 |
+
"loss": 1.0826,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7389787503964478,
|
| 1644 |
+
"grad_norm": 0.28126577459530283,
|
| 1645 |
+
"learning_rate": 0.00019335394401911082,
|
| 1646 |
+
"loss": 1.0719,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7421503330161751,
|
| 1651 |
+
"grad_norm": 0.2650849332335025,
|
| 1652 |
+
"learning_rate": 0.00018899801295547476,
|
| 1653 |
+
"loss": 1.0742,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7453219156359023,
|
| 1658 |
+
"grad_norm": 0.2603829852111257,
|
| 1659 |
+
"learning_rate": 0.00018468024520007764,
|
| 1660 |
+
"loss": 1.0772,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7484934982556296,
|
| 1665 |
+
"grad_norm": 0.2527087543783394,
|
| 1666 |
+
"learning_rate": 0.00018040117058973316,
|
| 1667 |
+
"loss": 1.0595,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7516650808753568,
|
| 1672 |
+
"grad_norm": 0.24678722431639855,
|
| 1673 |
+
"learning_rate": 0.0001761613142131867,
|
| 1674 |
+
"loss": 1.0469,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.754836663495084,
|
| 1679 |
+
"grad_norm": 0.25910814410326344,
|
| 1680 |
+
"learning_rate": 0.00017196119634668293,
|
| 1681 |
+
"loss": 1.0627,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7580082461148113,
|
| 1686 |
+
"grad_norm": 0.26173306347429054,
|
| 1687 |
+
"learning_rate": 0.00016780133239012075,
|
| 1688 |
+
"loss": 1.0607,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7611798287345385,
|
| 1693 |
+
"grad_norm": 0.24651016867032868,
|
| 1694 |
+
"learning_rate": 0.0001636822328038095,
|
| 1695 |
+
"loss": 1.0546,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7643514113542658,
|
| 1700 |
+
"grad_norm": 0.28020957707447064,
|
| 1701 |
+
"learning_rate": 0.00015960440304582858,
|
| 1702 |
+
"loss": 1.0579,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.767522993973993,
|
| 1707 |
+
"grad_norm": 0.26371476098524943,
|
| 1708 |
+
"learning_rate": 0.00015556834351000354,
|
| 1709 |
+
"loss": 1.0537,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7706945765937203,
|
| 1714 |
+
"grad_norm": 0.24623457163199874,
|
| 1715 |
+
"learning_rate": 0.0001515745494645019,
|
| 1716 |
+
"loss": 1.045,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7738661592134475,
|
| 1721 |
+
"grad_norm": 0.3155558400199454,
|
| 1722 |
+
"learning_rate": 0.0001476235109910576,
|
| 1723 |
+
"loss": 1.0405,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7770377418331748,
|
| 1728 |
+
"grad_norm": 0.2789622570986826,
|
| 1729 |
+
"learning_rate": 0.00014371571292483393,
|
| 1730 |
+
"loss": 1.0381,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.780209324452902,
|
| 1735 |
+
"grad_norm": 0.2409114498230053,
|
| 1736 |
+
"learning_rate": 0.0001398516347949284,
|
| 1737 |
+
"loss": 1.0394,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7833809070726292,
|
| 1742 |
+
"grad_norm": 0.267235707838601,
|
| 1743 |
+
"learning_rate": 0.0001360317507655293,
|
| 1744 |
+
"loss": 1.0278,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7865524896923565,
|
| 1749 |
+
"grad_norm": 0.28458374786381546,
|
| 1750 |
+
"learning_rate": 0.00013225652957773044,
|
| 1751 |
+
"loss": 1.0326,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7897240723120837,
|
| 1756 |
+
"grad_norm": 0.25695712786415686,
|
| 1757 |
+
"learning_rate": 0.00012852643449201212,
|
| 1758 |
+
"loss": 1.023,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.792895654931811,
|
| 1763 |
+
"grad_norm": 0.2590457354954553,
|
| 1764 |
+
"learning_rate": 0.0001248419232313938,
|
| 1765 |
+
"loss": 1.0232,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7960672375515382,
|
| 1770 |
+
"grad_norm": 0.2715843775728456,
|
| 1771 |
+
"learning_rate": 0.000121203447925266,
|
| 1772 |
+
"loss": 1.0287,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7992388201712655,
|
| 1777 |
+
"grad_norm": 0.2398511137856279,
|
| 1778 |
+
"learning_rate": 0.00011761145505391024,
|
| 1779 |
+
"loss": 1.0186,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8024104027909927,
|
| 1784 |
+
"grad_norm": 0.27281371167245233,
|
| 1785 |
+
"learning_rate": 0.00011406638539370979,
|
| 1786 |
+
"loss": 1.0224,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8055819854107199,
|
| 1791 |
+
"grad_norm": 0.3188098317762095,
|
| 1792 |
+
"learning_rate": 0.00011056867396306292,
|
| 1793 |
+
"loss": 1.0092,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8087535680304472,
|
| 1798 |
+
"grad_norm": 0.3265540754130617,
|
| 1799 |
+
"learning_rate": 0.00010711874996900023,
|
| 1800 |
+
"loss": 1.0104,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8119251506501745,
|
| 1805 |
+
"grad_norm": 0.2607401452606644,
|
| 1806 |
+
"learning_rate": 0.00010371703675451733,
|
| 1807 |
+
"loss": 1.0114,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8150967332699017,
|
| 1812 |
+
"grad_norm": 0.2883090335939891,
|
| 1813 |
+
"learning_rate": 0.0001003639517466256,
|
| 1814 |
+
"loss": 1.0093,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8182683158896289,
|
| 1819 |
+
"grad_norm": 0.25562259108760305,
|
| 1820 |
+
"learning_rate": 9.705990640512907e-05,
|
| 1821 |
+
"loss": 0.9938,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8214398985093562,
|
| 1826 |
+
"grad_norm": 0.2753573095600564,
|
| 1827 |
+
"learning_rate": 9.380530617213456e-05,
|
| 1828 |
+
"loss": 1.0114,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8246114811290834,
|
| 1833 |
+
"grad_norm": 0.23998112779723507,
|
| 1834 |
+
"learning_rate": 9.060055042229881e-05,
|
| 1835 |
+
"loss": 1.0089,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8277830637488106,
|
| 1840 |
+
"grad_norm": 0.2524204007518801,
|
| 1841 |
+
"learning_rate": 8.74460324138216e-05,
|
| 1842 |
+
"loss": 1.007,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8309546463685379,
|
| 1847 |
+
"grad_norm": 0.2526736715480949,
|
| 1848 |
+
"learning_rate": 8.434213924018835e-05,
|
| 1849 |
+
"loss": 1.0,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8341262289882652,
|
| 1854 |
+
"grad_norm": 0.2503643121035892,
|
| 1855 |
+
"learning_rate": 8.128925178266927e-05,
|
| 1856 |
+
"loss": 0.9965,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8372978116079924,
|
| 1861 |
+
"grad_norm": 0.23244697445873563,
|
| 1862 |
+
"learning_rate": 7.828774466358179e-05,
|
| 1863 |
+
"loss": 0.9988,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8404693942277196,
|
| 1868 |
+
"grad_norm": 0.2560633353876119,
|
| 1869 |
+
"learning_rate": 7.53379862003195e-05,
|
| 1870 |
+
"loss": 1.0048,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8436409768474469,
|
| 1875 |
+
"grad_norm": 0.2421969390657872,
|
| 1876 |
+
"learning_rate": 7.244033836015695e-05,
|
| 1877 |
+
"loss": 0.9844,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8468125594671742,
|
| 1882 |
+
"grad_norm": 0.2607793031238756,
|
| 1883 |
+
"learning_rate": 6.95951567158305e-05,
|
| 1884 |
+
"loss": 0.9778,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8499841420869013,
|
| 1889 |
+
"grad_norm": 0.2698408656759126,
|
| 1890 |
+
"learning_rate": 6.680279040190746e-05,
|
| 1891 |
+
"loss": 0.9828,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8531557247066286,
|
| 1896 |
+
"grad_norm": 0.23841421968692497,
|
| 1897 |
+
"learning_rate": 6.406358207194224e-05,
|
| 1898 |
+
"loss": 0.9991,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8563273073263559,
|
| 1903 |
+
"grad_norm": 0.28084531889088754,
|
| 1904 |
+
"learning_rate": 6.137786785642985e-05,
|
| 1905 |
+
"loss": 0.9855,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8594988899460831,
|
| 1910 |
+
"grad_norm": 0.24806562901660065,
|
| 1911 |
+
"learning_rate": 5.8745977321558786e-05,
|
| 1912 |
+
"loss": 0.9747,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8626704725658103,
|
| 1917 |
+
"grad_norm": 0.2511271472454086,
|
| 1918 |
+
"learning_rate": 5.616823342876931e-05,
|
| 1919 |
+
"loss": 0.9758,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8658420551855376,
|
| 1924 |
+
"grad_norm": 0.24241834259427655,
|
| 1925 |
+
"learning_rate": 5.364495249512336e-05,
|
| 1926 |
+
"loss": 0.9765,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8690136378052649,
|
| 1931 |
+
"grad_norm": 0.23883983545121304,
|
| 1932 |
+
"learning_rate": 5.11764441544883e-05,
|
| 1933 |
+
"loss": 0.9808,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.872185220424992,
|
| 1938 |
+
"grad_norm": 0.24435079120648112,
|
| 1939 |
+
"learning_rate": 4.8763011319542025e-05,
|
| 1940 |
+
"loss": 0.9777,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8753568030447193,
|
| 1945 |
+
"grad_norm": 0.2629272614210174,
|
| 1946 |
+
"learning_rate": 4.6404950144602e-05,
|
| 1947 |
+
"loss": 0.9819,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8785283856644466,
|
| 1952 |
+
"grad_norm": 0.2663926565440222,
|
| 1953 |
+
"learning_rate": 4.4102549989283756e-05,
|
| 1954 |
+
"loss": 0.9675,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8816999682841739,
|
| 1959 |
+
"grad_norm": 0.2431315966754426,
|
| 1960 |
+
"learning_rate": 4.1856093382994067e-05,
|
| 1961 |
+
"loss": 0.9617,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.884871550903901,
|
| 1966 |
+
"grad_norm": 0.2615985135082817,
|
| 1967 |
+
"learning_rate": 3.966585599026051e-05,
|
| 1968 |
+
"loss": 0.9705,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8880431335236283,
|
| 1973 |
+
"grad_norm": 0.2437571495377507,
|
| 1974 |
+
"learning_rate": 3.753210657690537e-05,
|
| 1975 |
+
"loss": 0.9637,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8912147161433556,
|
| 1980 |
+
"grad_norm": 0.2867683320851404,
|
| 1981 |
+
"learning_rate": 3.5455106977064555e-05,
|
| 1982 |
+
"loss": 0.9813,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8943862987630827,
|
| 1987 |
+
"grad_norm": 0.23596873266775822,
|
| 1988 |
+
"learning_rate": 3.343511206105804e-05,
|
| 1989 |
+
"loss": 0.9654,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.89755788138281,
|
| 1994 |
+
"grad_norm": 0.2691994741352151,
|
| 1995 |
+
"learning_rate": 3.147236970411449e-05,
|
| 1996 |
+
"loss": 0.9559,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9007294640025373,
|
| 2001 |
+
"grad_norm": 0.25848494724563204,
|
| 2002 |
+
"learning_rate": 2.9567120755953858e-05,
|
| 2003 |
+
"loss": 0.9631,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9039010466222646,
|
| 2008 |
+
"grad_norm": 0.2329147999513356,
|
| 2009 |
+
"learning_rate": 2.7719599011233333e-05,
|
| 2010 |
+
"loss": 0.9654,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9070726292419917,
|
| 2015 |
+
"grad_norm": 0.24897956262619586,
|
| 2016 |
+
"learning_rate": 2.593003118085746e-05,
|
| 2017 |
+
"loss": 0.9686,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.910244211861719,
|
| 2022 |
+
"grad_norm": 0.24140747398754628,
|
| 2023 |
+
"learning_rate": 2.4198636864158684e-05,
|
| 2024 |
+
"loss": 0.9709,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.9134157944814463,
|
| 2029 |
+
"grad_norm": 0.23859470642150962,
|
| 2030 |
+
"learning_rate": 2.2525628521949837e-05,
|
| 2031 |
+
"loss": 0.9723,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9165873771011734,
|
| 2036 |
+
"grad_norm": 0.23614657553272203,
|
| 2037 |
+
"learning_rate": 2.091121145045327e-05,
|
| 2038 |
+
"loss": 0.96,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9197589597209007,
|
| 2043 |
+
"grad_norm": 0.2305582486421353,
|
| 2044 |
+
"learning_rate": 1.9355583756108407e-05,
|
| 2045 |
+
"loss": 0.9622,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.922930542340628,
|
| 2050 |
+
"grad_norm": 0.23818404259591164,
|
| 2051 |
+
"learning_rate": 1.7858936331262122e-05,
|
| 2052 |
+
"loss": 0.9612,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9261021249603553,
|
| 2057 |
+
"grad_norm": 0.2378653986328734,
|
| 2058 |
+
"learning_rate": 1.6421452830744365e-05,
|
| 2059 |
+
"loss": 0.9716,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9292737075800824,
|
| 2064 |
+
"grad_norm": 0.2329586634857264,
|
| 2065 |
+
"learning_rate": 1.5043309649331205e-05,
|
| 2066 |
+
"loss": 0.9558,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9324452901998097,
|
| 2071 |
+
"grad_norm": 0.4063284757979328,
|
| 2072 |
+
"learning_rate": 1.3724675900099959e-05,
|
| 2073 |
+
"loss": 0.9654,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.935616872819537,
|
| 2078 |
+
"grad_norm": 0.24883789905797735,
|
| 2079 |
+
"learning_rate": 1.246571339367658e-05,
|
| 2080 |
+
"loss": 0.9603,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9387884554392641,
|
| 2085 |
+
"grad_norm": 0.2484198570399255,
|
| 2086 |
+
"learning_rate": 1.1266576618380097e-05,
|
| 2087 |
+
"loss": 0.9579,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9419600380589914,
|
| 2092 |
+
"grad_norm": 0.24729636213521072,
|
| 2093 |
+
"learning_rate": 1.0127412721265218e-05,
|
| 2094 |
+
"loss": 0.9675,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.9451316206787187,
|
| 2099 |
+
"grad_norm": 0.2505398676306425,
|
| 2100 |
+
"learning_rate": 9.048361490065548e-06,
|
| 2101 |
+
"loss": 0.9526,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.948303203298446,
|
| 2106 |
+
"grad_norm": 0.26466516239326676,
|
| 2107 |
+
"learning_rate": 8.029555336040383e-06,
|
| 2108 |
+
"loss": 0.9661,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9514747859181731,
|
| 2113 |
+
"grad_norm": 0.24164934386910944,
|
| 2114 |
+
"learning_rate": 7.071119277726301e-06,
|
| 2115 |
+
"loss": 0.9577,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9546463685379004,
|
| 2120 |
+
"grad_norm": 0.23732953651039135,
|
| 2121 |
+
"learning_rate": 6.17317092559605e-06,
|
| 2122 |
+
"loss": 0.9562,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9578179511576277,
|
| 2127 |
+
"grad_norm": 0.23388339171880254,
|
| 2128 |
+
"learning_rate": 5.335820467626485e-06,
|
| 2129 |
+
"loss": 0.973,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9609895337773549,
|
| 2134 |
+
"grad_norm": 0.2318159543809297,
|
| 2135 |
+
"learning_rate": 4.559170655777267e-06,
|
| 2136 |
+
"loss": 0.9478,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9641611163970821,
|
| 2141 |
+
"grad_norm": 0.24710746850544488,
|
| 2142 |
+
"learning_rate": 3.843316793382123e-06,
|
| 2143 |
+
"loss": 0.9707,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9673326990168094,
|
| 2148 |
+
"grad_norm": 0.2862292656525582,
|
| 2149 |
+
"learning_rate": 3.188346723454083e-06,
|
| 2150 |
+
"loss": 0.9643,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9705042816365367,
|
| 2155 |
+
"grad_norm": 0.2380821089662019,
|
| 2156 |
+
"learning_rate": 2.594340817906271e-06,
|
| 2157 |
+
"loss": 0.9624,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9736758642562638,
|
| 2162 |
+
"grad_norm": 0.23933390384190942,
|
| 2163 |
+
"learning_rate": 2.0613719676891853e-06,
|
| 2164 |
+
"loss": 0.9599,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9768474468759911,
|
| 2169 |
+
"grad_norm": 0.24854365220185978,
|
| 2170 |
+
"learning_rate": 1.5895055738465169e-06,
|
| 2171 |
+
"loss": 0.9592,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9800190294957184,
|
| 2176 |
+
"grad_norm": 0.24336558825284596,
|
| 2177 |
+
"learning_rate": 1.1787995394893502e-06,
|
| 2178 |
+
"loss": 0.962,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9831906121154456,
|
| 2183 |
+
"grad_norm": 0.22749162744166387,
|
| 2184 |
+
"learning_rate": 8.293042626912328e-07,
|
| 2185 |
+
"loss": 0.9573,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9863621947351728,
|
| 2190 |
+
"grad_norm": 0.2390082283281969,
|
| 2191 |
+
"learning_rate": 5.410626303034017e-07,
|
| 2192 |
+
"loss": 0.9556,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9895337773549001,
|
| 2197 |
+
"grad_norm": 0.23360184137608606,
|
| 2198 |
+
"learning_rate": 3.141100126923813e-07,
|
| 2199 |
+
"loss": 0.9571,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9927053599746274,
|
| 2204 |
+
"grad_norm": 0.2367998742996814,
|
| 2205 |
+
"learning_rate": 1.4847425939956693e-07,
|
| 2206 |
+
"loss": 0.9495,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9958769425943546,
|
| 2211 |
+
"grad_norm": 0.2269345672235601,
|
| 2212 |
+
"learning_rate": 4.417569572368052e-08,
|
| 2213 |
+
"loss": 0.9499,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9990485252140818,
|
| 2218 |
+
"grad_norm": 0.2484799432231265,
|
| 2219 |
+
"learning_rate": 1.2271202268210324e-09,
|
| 2220 |
+
"loss": 0.956,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9996828417380272,
|
| 2225 |
+
"eval_loss": 2.245941638946533,
|
| 2226 |
+
"eval_runtime": 8.442,
|
| 2227 |
+
"eval_samples_per_second": 46.316,
|
| 2228 |
+
"eval_steps_per_second": 5.804,
|
| 2229 |
+
"step": 1576
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.9996828417380272,
|
| 2233 |
+
"step": 1576,
|
| 2234 |
+
"total_flos": 38663670988800.0,
|
| 2235 |
+
"train_loss": 1.4483577938854393,
|
| 2236 |
+
"train_runtime": 3409.163,
|
| 2237 |
+
"train_samples_per_second": 14.797,
|
| 2238 |
+
"train_steps_per_second": 0.462
|
| 2239 |
+
}
|
| 2240 |
+
],
|
| 2241 |
+
"logging_steps": 5,
|
| 2242 |
+
"max_steps": 1576,
|
| 2243 |
+
"num_input_tokens_seen": 0,
|
| 2244 |
+
"num_train_epochs": 1,
|
| 2245 |
+
"save_steps": 500,
|
| 2246 |
+
"stateful_callbacks": {
|
| 2247 |
+
"TrainerControl": {
|
| 2248 |
+
"args": {
|
| 2249 |
+
"should_epoch_stop": false,
|
| 2250 |
+
"should_evaluate": false,
|
| 2251 |
+
"should_log": false,
|
| 2252 |
+
"should_save": false,
|
| 2253 |
+
"should_training_stop": false
|
| 2254 |
+
},
|
| 2255 |
+
"attributes": {}
|
| 2256 |
+
}
|
| 2257 |
+
},
|
| 2258 |
+
"total_flos": 38663670988800.0,
|
| 2259 |
+
"train_batch_size": 4,
|
| 2260 |
+
"trial_name": null,
|
| 2261 |
+
"trial_params": null
|
| 2262 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7298d47d0acd3a88973e85468ff24e30141d19bdc3a91d7485f2b0cd219f18f
|
| 3 |
+
size 6840
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|