Upload folder using huggingface_hub
Browse files- added_tokens.json +5 -0
- config.json +43 -0
- generation_config.json +15 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +20 -0
- tokenizer_config.json +44 -0
- trainer_state.json +2002 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./checkpoint/llmga-Qwen2-0.5B-full-finetune",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaQwen2ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"freeze_mm_mlp_adapter": false,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 896,
|
| 12 |
+
"image_aspect_ratio": "resizesquare",
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4864,
|
| 15 |
+
"max_position_embeddings": 32768,
|
| 16 |
+
"max_window_layers": 24,
|
| 17 |
+
"mm_hidden_size": 1024,
|
| 18 |
+
"mm_patch_merge_type": "flat",
|
| 19 |
+
"mm_projector_lr": null,
|
| 20 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 21 |
+
"mm_use_im_patch_token": false,
|
| 22 |
+
"mm_use_im_start_end": false,
|
| 23 |
+
"mm_vision_select_feature": "patch",
|
| 24 |
+
"mm_vision_select_layer": -2,
|
| 25 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 26 |
+
"model_type": "llava_Qwen2",
|
| 27 |
+
"num_attention_heads": 14,
|
| 28 |
+
"num_hidden_layers": 24,
|
| 29 |
+
"num_key_value_heads": 2,
|
| 30 |
+
"rms_norm_eps": 1e-06,
|
| 31 |
+
"rope_theta": 1000000.0,
|
| 32 |
+
"sliding_window": 32768,
|
| 33 |
+
"tie_word_embeddings": true,
|
| 34 |
+
"tokenizer_model_max_length": 2048,
|
| 35 |
+
"tokenizer_padding_side": "right",
|
| 36 |
+
"torch_dtype": "bfloat16",
|
| 37 |
+
"transformers_version": "4.41.2",
|
| 38 |
+
"tune_mm_mlp_adapter": false,
|
| 39 |
+
"use_cache": true,
|
| 40 |
+
"use_mm_proj": true,
|
| 41 |
+
"use_sliding_window": false,
|
| 42 |
+
"vocab_size": 151936
|
| 43 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"attn_implementation": "flash_attention_2",
|
| 3 |
+
"bos_token_id": 151643,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
151645,
|
| 7 |
+
151643
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 151643,
|
| 10 |
+
"repetition_penalty": 1.1,
|
| 11 |
+
"temperature": 0.7,
|
| 12 |
+
"top_k": 20,
|
| 13 |
+
"top_p": 0.8,
|
| 14 |
+
"transformers_version": "4.41.2"
|
| 15 |
+
}
|
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 sha256:9413f988a78ccea3c8e711e3f01656c26d2f61cd0e284a4c8a68079559357a95
|
| 3 |
+
size 1598618728
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>"
|
| 5 |
+
],
|
| 6 |
+
"eos_token": {
|
| 7 |
+
"content": "<|im_end|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"pad_token": {
|
| 14 |
+
"content": "<|endoftext|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
}
|
| 20 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"additional_special_tokens": [
|
| 30 |
+
"<|im_start|>",
|
| 31 |
+
"<|im_end|>"
|
| 32 |
+
],
|
| 33 |
+
"bos_token": null,
|
| 34 |
+
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 35 |
+
"clean_up_tokenization_spaces": false,
|
| 36 |
+
"eos_token": "<|im_end|>",
|
| 37 |
+
"errors": "replace",
|
| 38 |
+
"model_max_length": 2048,
|
| 39 |
+
"pad_token": "<|endoftext|>",
|
| 40 |
+
"padding_side": "right",
|
| 41 |
+
"split_special_tokens": false,
|
| 42 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 43 |
+
"unk_token": null
|
| 44 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2002 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 280,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0035714285714285713,
|
| 13 |
+
"grad_norm": 4.908648608990839,
|
| 14 |
+
"learning_rate": 2.222222222222222e-06,
|
| 15 |
+
"loss": 1.7961,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007142857142857143,
|
| 20 |
+
"grad_norm": 4.592180104184582,
|
| 21 |
+
"learning_rate": 4.444444444444444e-06,
|
| 22 |
+
"loss": 1.713,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010714285714285714,
|
| 27 |
+
"grad_norm": 4.69308510471444,
|
| 28 |
+
"learning_rate": 6.666666666666667e-06,
|
| 29 |
+
"loss": 1.6831,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014285714285714285,
|
| 34 |
+
"grad_norm": 3.653347608269976,
|
| 35 |
+
"learning_rate": 8.888888888888888e-06,
|
| 36 |
+
"loss": 1.5785,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.017857142857142856,
|
| 41 |
+
"grad_norm": 2.804649047110185,
|
| 42 |
+
"learning_rate": 1.1111111111111113e-05,
|
| 43 |
+
"loss": 1.5988,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02142857142857143,
|
| 48 |
+
"grad_norm": 3.1675123269741516,
|
| 49 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 50 |
+
"loss": 1.4795,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.025,
|
| 55 |
+
"grad_norm": 3.6958086333820566,
|
| 56 |
+
"learning_rate": 1.555555555555556e-05,
|
| 57 |
+
"loss": 1.5327,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02857142857142857,
|
| 62 |
+
"grad_norm": 5.724480820109564,
|
| 63 |
+
"learning_rate": 1.7777777777777777e-05,
|
| 64 |
+
"loss": 1.4916,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03214285714285714,
|
| 69 |
+
"grad_norm": 4.073008923111407,
|
| 70 |
+
"learning_rate": 2e-05,
|
| 71 |
+
"loss": 1.4973,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03571428571428571,
|
| 76 |
+
"grad_norm": 2.6759891152038526,
|
| 77 |
+
"learning_rate": 1.9999328066483867e-05,
|
| 78 |
+
"loss": 1.3998,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.039285714285714285,
|
| 83 |
+
"grad_norm": 3.2603665327453277,
|
| 84 |
+
"learning_rate": 1.9997312356234385e-05,
|
| 85 |
+
"loss": 1.3604,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04285714285714286,
|
| 90 |
+
"grad_norm": 3.6130206264673457,
|
| 91 |
+
"learning_rate": 1.999395314013622e-05,
|
| 92 |
+
"loss": 1.3048,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04642857142857143,
|
| 97 |
+
"grad_norm": 2.661927084764187,
|
| 98 |
+
"learning_rate": 1.998925086962334e-05,
|
| 99 |
+
"loss": 1.2912,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05,
|
| 104 |
+
"grad_norm": 2.576973416368585,
|
| 105 |
+
"learning_rate": 1.998320617661839e-05,
|
| 106 |
+
"loss": 1.2325,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05357142857142857,
|
| 111 |
+
"grad_norm": 2.4728531527823385,
|
| 112 |
+
"learning_rate": 1.997581987344772e-05,
|
| 113 |
+
"loss": 1.2227,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05714285714285714,
|
| 118 |
+
"grad_norm": 2.6919633979798023,
|
| 119 |
+
"learning_rate": 1.9967092952732266e-05,
|
| 120 |
+
"loss": 1.1903,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.060714285714285714,
|
| 125 |
+
"grad_norm": 2.8589949375920414,
|
| 126 |
+
"learning_rate": 1.9957026587254136e-05,
|
| 127 |
+
"loss": 1.1843,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06428571428571428,
|
| 132 |
+
"grad_norm": 2.3469600295209823,
|
| 133 |
+
"learning_rate": 1.9945622129799e-05,
|
| 134 |
+
"loss": 1.1089,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06785714285714285,
|
| 139 |
+
"grad_norm": 2.5116022382668968,
|
| 140 |
+
"learning_rate": 1.9932881112974298e-05,
|
| 141 |
+
"loss": 1.0887,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07142857142857142,
|
| 146 |
+
"grad_norm": 2.2064320702567413,
|
| 147 |
+
"learning_rate": 1.9918805249003272e-05,
|
| 148 |
+
"loss": 1.0124,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.075,
|
| 153 |
+
"grad_norm": 2.5511459947910557,
|
| 154 |
+
"learning_rate": 1.9903396429494882e-05,
|
| 155 |
+
"loss": 0.9848,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07857142857142857,
|
| 160 |
+
"grad_norm": 2.074497064278477,
|
| 161 |
+
"learning_rate": 1.9886656725189575e-05,
|
| 162 |
+
"loss": 0.9706,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08214285714285714,
|
| 167 |
+
"grad_norm": 2.0029563935827714,
|
| 168 |
+
"learning_rate": 1.9868588385681035e-05,
|
| 169 |
+
"loss": 0.9629,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08571428571428572,
|
| 174 |
+
"grad_norm": 2.166398531324486,
|
| 175 |
+
"learning_rate": 1.9849193839113833e-05,
|
| 176 |
+
"loss": 0.8597,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08928571428571429,
|
| 181 |
+
"grad_norm": 2.034388942107051,
|
| 182 |
+
"learning_rate": 1.9828475691857148e-05,
|
| 183 |
+
"loss": 0.8356,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09285714285714286,
|
| 188 |
+
"grad_norm": 1.8616897229035658,
|
| 189 |
+
"learning_rate": 1.9806436728154484e-05,
|
| 190 |
+
"loss": 0.7967,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.09642857142857143,
|
| 195 |
+
"grad_norm": 1.937793646144307,
|
| 196 |
+
"learning_rate": 1.9783079909749516e-05,
|
| 197 |
+
"loss": 0.7902,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.1,
|
| 202 |
+
"grad_norm": 2.03802418267336,
|
| 203 |
+
"learning_rate": 1.975840837548807e-05,
|
| 204 |
+
"loss": 0.8302,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10357142857142858,
|
| 209 |
+
"grad_norm": 1.7299975214673873,
|
| 210 |
+
"learning_rate": 1.9732425440896298e-05,
|
| 211 |
+
"loss": 0.8622,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.10714285714285714,
|
| 216 |
+
"grad_norm": 1.806743882141901,
|
| 217 |
+
"learning_rate": 1.9705134597735113e-05,
|
| 218 |
+
"loss": 0.7735,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11071428571428571,
|
| 223 |
+
"grad_norm": 1.8544287022749248,
|
| 224 |
+
"learning_rate": 1.9676539513530967e-05,
|
| 225 |
+
"loss": 0.6243,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11428571428571428,
|
| 230 |
+
"grad_norm": 1.8369289759240388,
|
| 231 |
+
"learning_rate": 1.9646644031082948e-05,
|
| 232 |
+
"loss": 0.6859,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.11785714285714285,
|
| 237 |
+
"grad_norm": 1.941599747726953,
|
| 238 |
+
"learning_rate": 1.9615452167946383e-05,
|
| 239 |
+
"loss": 0.7323,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12142857142857143,
|
| 244 |
+
"grad_norm": 1.9016904671057466,
|
| 245 |
+
"learning_rate": 1.958296811589293e-05,
|
| 246 |
+
"loss": 0.6924,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.125,
|
| 251 |
+
"grad_norm": 2.0994171001442754,
|
| 252 |
+
"learning_rate": 1.954919624034725e-05,
|
| 253 |
+
"loss": 0.7021,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.12857142857142856,
|
| 258 |
+
"grad_norm": 1.8461885488185878,
|
| 259 |
+
"learning_rate": 1.951414107980036e-05,
|
| 260 |
+
"loss": 0.6776,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13214285714285715,
|
| 265 |
+
"grad_norm": 1.697750537080288,
|
| 266 |
+
"learning_rate": 1.9477807345199717e-05,
|
| 267 |
+
"loss": 0.7172,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1357142857142857,
|
| 272 |
+
"grad_norm": 1.8267671959875706,
|
| 273 |
+
"learning_rate": 1.9440199919316125e-05,
|
| 274 |
+
"loss": 0.6439,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1392857142857143,
|
| 279 |
+
"grad_norm": 1.6993602930346685,
|
| 280 |
+
"learning_rate": 1.9401323856087573e-05,
|
| 281 |
+
"loss": 0.618,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14285714285714285,
|
| 286 |
+
"grad_norm": 1.6727471871735518,
|
| 287 |
+
"learning_rate": 1.936118437994003e-05,
|
| 288 |
+
"loss": 0.6352,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.14642857142857144,
|
| 293 |
+
"grad_norm": 1.754512057986924,
|
| 294 |
+
"learning_rate": 1.9319786885085366e-05,
|
| 295 |
+
"loss": 0.5894,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15,
|
| 300 |
+
"grad_norm": 1.740151549924287,
|
| 301 |
+
"learning_rate": 1.927713693479643e-05,
|
| 302 |
+
"loss": 0.6899,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15357142857142858,
|
| 307 |
+
"grad_norm": 1.594981649980205,
|
| 308 |
+
"learning_rate": 1.923324026065944e-05,
|
| 309 |
+
"loss": 0.5648,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.15714285714285714,
|
| 314 |
+
"grad_norm": 1.5576937266056854,
|
| 315 |
+
"learning_rate": 1.918810276180372e-05,
|
| 316 |
+
"loss": 0.5596,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16071428571428573,
|
| 321 |
+
"grad_norm": 1.7216253449123322,
|
| 322 |
+
"learning_rate": 1.9141730504108923e-05,
|
| 323 |
+
"loss": 0.6042,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16428571428571428,
|
| 328 |
+
"grad_norm": 1.6040787546322597,
|
| 329 |
+
"learning_rate": 1.9094129719389886e-05,
|
| 330 |
+
"loss": 0.5168,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.16785714285714284,
|
| 335 |
+
"grad_norm": 1.731120413859831,
|
| 336 |
+
"learning_rate": 1.904530680455914e-05,
|
| 337 |
+
"loss": 0.6288,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17142857142857143,
|
| 342 |
+
"grad_norm": 1.626971149833354,
|
| 343 |
+
"learning_rate": 1.8995268320767254e-05,
|
| 344 |
+
"loss": 0.5288,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.175,
|
| 349 |
+
"grad_norm": 1.4082904642307315,
|
| 350 |
+
"learning_rate": 1.894402099252109e-05,
|
| 351 |
+
"loss": 0.5477,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.17857142857142858,
|
| 356 |
+
"grad_norm": 1.6671028870240991,
|
| 357 |
+
"learning_rate": 1.889157170678015e-05,
|
| 358 |
+
"loss": 0.5541,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18214285714285713,
|
| 363 |
+
"grad_norm": 1.629789549819335,
|
| 364 |
+
"learning_rate": 1.8837927512031022e-05,
|
| 365 |
+
"loss": 0.4901,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.18571428571428572,
|
| 370 |
+
"grad_norm": 1.565700755496585,
|
| 371 |
+
"learning_rate": 1.8783095617340193e-05,
|
| 372 |
+
"loss": 0.4624,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.18928571428571428,
|
| 377 |
+
"grad_norm": 1.4678500727905517,
|
| 378 |
+
"learning_rate": 1.872708339138522e-05,
|
| 379 |
+
"loss": 0.4271,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19285714285714287,
|
| 384 |
+
"grad_norm": 1.4609229980872802,
|
| 385 |
+
"learning_rate": 1.866989836146449e-05,
|
| 386 |
+
"loss": 0.4376,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.19642857142857142,
|
| 391 |
+
"grad_norm": 1.6797655019359938,
|
| 392 |
+
"learning_rate": 1.861154821248565e-05,
|
| 393 |
+
"loss": 0.5878,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2,
|
| 398 |
+
"grad_norm": 1.577613646211982,
|
| 399 |
+
"learning_rate": 1.8552040785932846e-05,
|
| 400 |
+
"loss": 0.4783,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.20357142857142857,
|
| 405 |
+
"grad_norm": 1.7027221419545615,
|
| 406 |
+
"learning_rate": 1.849138407881296e-05,
|
| 407 |
+
"loss": 0.3975,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.20714285714285716,
|
| 412 |
+
"grad_norm": 1.670046977724538,
|
| 413 |
+
"learning_rate": 1.8429586242580884e-05,
|
| 414 |
+
"loss": 0.4367,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.21071428571428572,
|
| 419 |
+
"grad_norm": 1.465983705393668,
|
| 420 |
+
"learning_rate": 1.8366655582044096e-05,
|
| 421 |
+
"loss": 0.3857,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21428571428571427,
|
| 426 |
+
"grad_norm": 1.619649836022893,
|
| 427 |
+
"learning_rate": 1.83026005542466e-05,
|
| 428 |
+
"loss": 0.4711,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.21785714285714286,
|
| 433 |
+
"grad_norm": 1.570708639647691,
|
| 434 |
+
"learning_rate": 1.8237429767332407e-05,
|
| 435 |
+
"loss": 0.4524,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22142857142857142,
|
| 440 |
+
"grad_norm": 1.476746454110955,
|
| 441 |
+
"learning_rate": 1.8171151979388715e-05,
|
| 442 |
+
"loss": 0.3994,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.225,
|
| 447 |
+
"grad_norm": 1.6497165796916549,
|
| 448 |
+
"learning_rate": 1.8103776097268942e-05,
|
| 449 |
+
"loss": 0.4169,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.22857142857142856,
|
| 454 |
+
"grad_norm": 1.30798054852489,
|
| 455 |
+
"learning_rate": 1.803531117539577e-05,
|
| 456 |
+
"loss": 0.385,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.23214285714285715,
|
| 461 |
+
"grad_norm": 1.3237517166357549,
|
| 462 |
+
"learning_rate": 1.7965766414544328e-05,
|
| 463 |
+
"loss": 0.348,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2357142857142857,
|
| 468 |
+
"grad_norm": 1.6055539374817416,
|
| 469 |
+
"learning_rate": 1.7895151160605758e-05,
|
| 470 |
+
"loss": 0.4277,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2392857142857143,
|
| 475 |
+
"grad_norm": 1.6951560958360459,
|
| 476 |
+
"learning_rate": 1.782347490333123e-05,
|
| 477 |
+
"loss": 0.4142,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.24285714285714285,
|
| 482 |
+
"grad_norm": 1.2191039708896538,
|
| 483 |
+
"learning_rate": 1.775074727505667e-05,
|
| 484 |
+
"loss": 0.2954,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.24642857142857144,
|
| 489 |
+
"grad_norm": 1.5607149280052393,
|
| 490 |
+
"learning_rate": 1.7676978049408262e-05,
|
| 491 |
+
"loss": 0.3828,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.25,
|
| 496 |
+
"grad_norm": 1.5038483401175802,
|
| 497 |
+
"learning_rate": 1.7602177139989046e-05,
|
| 498 |
+
"loss": 0.405,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.25357142857142856,
|
| 503 |
+
"grad_norm": 1.6682214568606215,
|
| 504 |
+
"learning_rate": 1.7526354599046637e-05,
|
| 505 |
+
"loss": 0.3953,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2571428571428571,
|
| 510 |
+
"grad_norm": 1.347036705570398,
|
| 511 |
+
"learning_rate": 1.7449520616122344e-05,
|
| 512 |
+
"loss": 0.3629,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.26071428571428573,
|
| 517 |
+
"grad_norm": 1.2414625150772796,
|
| 518 |
+
"learning_rate": 1.7371685516681825e-05,
|
| 519 |
+
"loss": 0.3161,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2642857142857143,
|
| 524 |
+
"grad_norm": 1.3436886422108463,
|
| 525 |
+
"learning_rate": 1.7292859760727493e-05,
|
| 526 |
+
"loss": 0.3269,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.26785714285714285,
|
| 531 |
+
"grad_norm": 1.3530913321365612,
|
| 532 |
+
"learning_rate": 1.721305394139282e-05,
|
| 533 |
+
"loss": 0.2728,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2714285714285714,
|
| 538 |
+
"grad_norm": 1.222920831545463,
|
| 539 |
+
"learning_rate": 1.7132278783518756e-05,
|
| 540 |
+
"loss": 0.2876,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.275,
|
| 545 |
+
"grad_norm": 1.5358149513341914,
|
| 546 |
+
"learning_rate": 1.7050545142212483e-05,
|
| 547 |
+
"loss": 0.3957,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2785714285714286,
|
| 552 |
+
"grad_norm": 1.4149548491141548,
|
| 553 |
+
"learning_rate": 1.696786400138859e-05,
|
| 554 |
+
"loss": 0.3104,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.28214285714285714,
|
| 559 |
+
"grad_norm": 1.7669148749254266,
|
| 560 |
+
"learning_rate": 1.6884246472293018e-05,
|
| 561 |
+
"loss": 0.3743,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2857142857142857,
|
| 566 |
+
"grad_norm": 1.265504102534398,
|
| 567 |
+
"learning_rate": 1.679970379200983e-05,
|
| 568 |
+
"loss": 0.317,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2892857142857143,
|
| 573 |
+
"grad_norm": 1.2888395180660637,
|
| 574 |
+
"learning_rate": 1.6714247321951106e-05,
|
| 575 |
+
"loss": 0.342,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.29285714285714287,
|
| 580 |
+
"grad_norm": 1.422200915278466,
|
| 581 |
+
"learning_rate": 1.6627888546330136e-05,
|
| 582 |
+
"loss": 0.4185,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.29642857142857143,
|
| 587 |
+
"grad_norm": 1.4661329628560202,
|
| 588 |
+
"learning_rate": 1.654063907061807e-05,
|
| 589 |
+
"loss": 0.3713,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3,
|
| 594 |
+
"grad_norm": 1.4678935830745146,
|
| 595 |
+
"learning_rate": 1.64525106199843e-05,
|
| 596 |
+
"loss": 0.3066,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.30357142857142855,
|
| 601 |
+
"grad_norm": 1.2825311493245366,
|
| 602 |
+
"learning_rate": 1.6363515037720774e-05,
|
| 603 |
+
"loss": 0.3134,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.30714285714285716,
|
| 608 |
+
"grad_norm": 1.3008599464968365,
|
| 609 |
+
"learning_rate": 1.6273664283650393e-05,
|
| 610 |
+
"loss": 0.3824,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3107142857142857,
|
| 615 |
+
"grad_norm": 1.5609153598311494,
|
| 616 |
+
"learning_rate": 1.6182970432519772e-05,
|
| 617 |
+
"loss": 0.2641,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3142857142857143,
|
| 622 |
+
"grad_norm": 1.321725446631543,
|
| 623 |
+
"learning_rate": 1.609144567237658e-05,
|
| 624 |
+
"loss": 0.3433,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.31785714285714284,
|
| 629 |
+
"grad_norm": 1.230389929756339,
|
| 630 |
+
"learning_rate": 1.5999102302931585e-05,
|
| 631 |
+
"loss": 0.2882,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.32142857142857145,
|
| 636 |
+
"grad_norm": 1.282470788355045,
|
| 637 |
+
"learning_rate": 1.5905952733905777e-05,
|
| 638 |
+
"loss": 0.4001,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.325,
|
| 643 |
+
"grad_norm": 1.244760100205609,
|
| 644 |
+
"learning_rate": 1.5812009483362643e-05,
|
| 645 |
+
"loss": 0.3303,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.32857142857142857,
|
| 650 |
+
"grad_norm": 1.3077612857336067,
|
| 651 |
+
"learning_rate": 1.5717285176025913e-05,
|
| 652 |
+
"loss": 0.3009,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.33214285714285713,
|
| 657 |
+
"grad_norm": 1.5007751444825745,
|
| 658 |
+
"learning_rate": 1.5621792541582968e-05,
|
| 659 |
+
"loss": 0.4037,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3357142857142857,
|
| 664 |
+
"grad_norm": 1.2889530196987555,
|
| 665 |
+
"learning_rate": 1.552554441297413e-05,
|
| 666 |
+
"loss": 0.3166,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3392857142857143,
|
| 671 |
+
"grad_norm": 1.471774386098153,
|
| 672 |
+
"learning_rate": 1.5428553724668103e-05,
|
| 673 |
+
"loss": 0.3141,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.34285714285714286,
|
| 678 |
+
"grad_norm": 1.2425633719745615,
|
| 679 |
+
"learning_rate": 1.533083351092372e-05,
|
| 680 |
+
"loss": 0.3032,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3464285714285714,
|
| 685 |
+
"grad_norm": 1.105289454224738,
|
| 686 |
+
"learning_rate": 1.5232396904038352e-05,
|
| 687 |
+
"loss": 0.2762,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.35,
|
| 692 |
+
"grad_norm": 1.2970218695315465,
|
| 693 |
+
"learning_rate": 1.5133257132583074e-05,
|
| 694 |
+
"loss": 0.3101,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3535714285714286,
|
| 699 |
+
"grad_norm": 1.3017528350777372,
|
| 700 |
+
"learning_rate": 1.503342751962493e-05,
|
| 701 |
+
"loss": 0.3731,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.35714285714285715,
|
| 706 |
+
"grad_norm": 1.3494066259729216,
|
| 707 |
+
"learning_rate": 1.4932921480936491e-05,
|
| 708 |
+
"loss": 0.2738,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3607142857142857,
|
| 713 |
+
"grad_norm": 1.2452542674259262,
|
| 714 |
+
"learning_rate": 1.4831752523192949e-05,
|
| 715 |
+
"loss": 0.2732,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.36428571428571427,
|
| 720 |
+
"grad_norm": 1.1826372513240704,
|
| 721 |
+
"learning_rate": 1.4729934242157005e-05,
|
| 722 |
+
"loss": 0.2772,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3678571428571429,
|
| 727 |
+
"grad_norm": 1.3655889492363933,
|
| 728 |
+
"learning_rate": 1.4627480320851775e-05,
|
| 729 |
+
"loss": 0.3117,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.37142857142857144,
|
| 734 |
+
"grad_norm": 1.0905245915838764,
|
| 735 |
+
"learning_rate": 1.4524404527721977e-05,
|
| 736 |
+
"loss": 0.2421,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.375,
|
| 741 |
+
"grad_norm": 1.1892908660572088,
|
| 742 |
+
"learning_rate": 1.4420720714783635e-05,
|
| 743 |
+
"loss": 0.3403,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.37857142857142856,
|
| 748 |
+
"grad_norm": 1.3612551275939382,
|
| 749 |
+
"learning_rate": 1.4316442815762543e-05,
|
| 750 |
+
"loss": 0.3111,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3821428571428571,
|
| 755 |
+
"grad_norm": 1.0293414732881423,
|
| 756 |
+
"learning_rate": 1.4211584844221771e-05,
|
| 757 |
+
"loss": 0.2198,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.38571428571428573,
|
| 762 |
+
"grad_norm": 1.0973339441554495,
|
| 763 |
+
"learning_rate": 1.4106160891678422e-05,
|
| 764 |
+
"loss": 0.2438,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3892857142857143,
|
| 769 |
+
"grad_norm": 0.9728591339955907,
|
| 770 |
+
"learning_rate": 1.4000185125709919e-05,
|
| 771 |
+
"loss": 0.2189,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.39285714285714285,
|
| 776 |
+
"grad_norm": 1.0719702208838044,
|
| 777 |
+
"learning_rate": 1.3893671788050073e-05,
|
| 778 |
+
"loss": 0.2782,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3964285714285714,
|
| 783 |
+
"grad_norm": 0.9706227479864399,
|
| 784 |
+
"learning_rate": 1.3786635192675184e-05,
|
| 785 |
+
"loss": 0.227,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4,
|
| 790 |
+
"grad_norm": 1.4196235328043891,
|
| 791 |
+
"learning_rate": 1.3679089723880427e-05,
|
| 792 |
+
"loss": 0.3279,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.4035714285714286,
|
| 797 |
+
"grad_norm": 0.9875713875583682,
|
| 798 |
+
"learning_rate": 1.35710498343468e-05,
|
| 799 |
+
"loss": 0.2223,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.40714285714285714,
|
| 804 |
+
"grad_norm": 1.2683114513213385,
|
| 805 |
+
"learning_rate": 1.3462530043198874e-05,
|
| 806 |
+
"loss": 0.3193,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.4107142857142857,
|
| 811 |
+
"grad_norm": 1.1548967936561843,
|
| 812 |
+
"learning_rate": 1.3353544934053618e-05,
|
| 813 |
+
"loss": 0.2418,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4142857142857143,
|
| 818 |
+
"grad_norm": 1.1578835473542461,
|
| 819 |
+
"learning_rate": 1.324410915306055e-05,
|
| 820 |
+
"loss": 0.2741,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.41785714285714287,
|
| 825 |
+
"grad_norm": 0.867810988436804,
|
| 826 |
+
"learning_rate": 1.3134237406933493e-05,
|
| 827 |
+
"loss": 0.2182,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.42142857142857143,
|
| 832 |
+
"grad_norm": 1.226765456766979,
|
| 833 |
+
"learning_rate": 1.3023944460974183e-05,
|
| 834 |
+
"loss": 0.2486,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.425,
|
| 839 |
+
"grad_norm": 1.2047266305046154,
|
| 840 |
+
"learning_rate": 1.2913245137088024e-05,
|
| 841 |
+
"loss": 0.2654,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.42857142857142855,
|
| 846 |
+
"grad_norm": 1.0309280957839033,
|
| 847 |
+
"learning_rate": 1.2802154311792196e-05,
|
| 848 |
+
"loss": 0.2194,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.43214285714285716,
|
| 853 |
+
"grad_norm": 1.1216683723429337,
|
| 854 |
+
"learning_rate": 1.2690686914216475e-05,
|
| 855 |
+
"loss": 0.2471,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.4357142857142857,
|
| 860 |
+
"grad_norm": 0.7992505679628088,
|
| 861 |
+
"learning_rate": 1.2578857924096935e-05,
|
| 862 |
+
"loss": 0.2053,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4392857142857143,
|
| 867 |
+
"grad_norm": 0.9521815104123159,
|
| 868 |
+
"learning_rate": 1.2466682369762883e-05,
|
| 869 |
+
"loss": 0.218,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.44285714285714284,
|
| 874 |
+
"grad_norm": 1.0131891476301753,
|
| 875 |
+
"learning_rate": 1.2354175326117252e-05,
|
| 876 |
+
"loss": 0.2797,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.44642857142857145,
|
| 881 |
+
"grad_norm": 1.1514536611512578,
|
| 882 |
+
"learning_rate": 1.2241351912610726e-05,
|
| 883 |
+
"loss": 0.326,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.45,
|
| 888 |
+
"grad_norm": 1.4037867589116553,
|
| 889 |
+
"learning_rate": 1.212822729120989e-05,
|
| 890 |
+
"loss": 0.2137,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.45357142857142857,
|
| 895 |
+
"grad_norm": 0.9592307976271547,
|
| 896 |
+
"learning_rate": 1.2014816664359671e-05,
|
| 897 |
+
"loss": 0.2489,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.45714285714285713,
|
| 902 |
+
"grad_norm": 1.239268404867852,
|
| 903 |
+
"learning_rate": 1.190113527294032e-05,
|
| 904 |
+
"loss": 0.3135,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4607142857142857,
|
| 909 |
+
"grad_norm": 0.9664072810151023,
|
| 910 |
+
"learning_rate": 1.178719839421925e-05,
|
| 911 |
+
"loss": 0.2025,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.4642857142857143,
|
| 916 |
+
"grad_norm": 1.000615478826764,
|
| 917 |
+
"learning_rate": 1.1673021339797967e-05,
|
| 918 |
+
"loss": 0.2701,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.46785714285714286,
|
| 923 |
+
"grad_norm": 1.0284186941827786,
|
| 924 |
+
"learning_rate": 1.15586194535544e-05,
|
| 925 |
+
"loss": 0.2448,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4714285714285714,
|
| 930 |
+
"grad_norm": 0.7308129021048215,
|
| 931 |
+
"learning_rate": 1.1444008109580884e-05,
|
| 932 |
+
"loss": 0.1569,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.475,
|
| 937 |
+
"grad_norm": 0.8601356644168414,
|
| 938 |
+
"learning_rate": 1.1329202710118088e-05,
|
| 939 |
+
"loss": 0.2752,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4785714285714286,
|
| 944 |
+
"grad_norm": 1.1174847463212987,
|
| 945 |
+
"learning_rate": 1.1214218683485159e-05,
|
| 946 |
+
"loss": 0.2813,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.48214285714285715,
|
| 951 |
+
"grad_norm": 0.9222986025138733,
|
| 952 |
+
"learning_rate": 1.1099071482006361e-05,
|
| 953 |
+
"loss": 0.2127,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4857142857142857,
|
| 958 |
+
"grad_norm": 1.0153753324402643,
|
| 959 |
+
"learning_rate": 1.0983776579934483e-05,
|
| 960 |
+
"loss": 0.3222,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.48928571428571427,
|
| 965 |
+
"grad_norm": 1.1498734203876837,
|
| 966 |
+
"learning_rate": 1.0868349471371316e-05,
|
| 967 |
+
"loss": 0.3267,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4928571428571429,
|
| 972 |
+
"grad_norm": 1.0327891450611766,
|
| 973 |
+
"learning_rate": 1.0752805668185442e-05,
|
| 974 |
+
"loss": 0.2247,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.49642857142857144,
|
| 979 |
+
"grad_norm": 1.04721921113995,
|
| 980 |
+
"learning_rate": 1.0637160697927651e-05,
|
| 981 |
+
"loss": 0.285,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5,
|
| 986 |
+
"grad_norm": 0.9226664091190546,
|
| 987 |
+
"learning_rate": 1.0521430101744238e-05,
|
| 988 |
+
"loss": 0.2138,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5035714285714286,
|
| 993 |
+
"grad_norm": 1.2238486576835246,
|
| 994 |
+
"learning_rate": 1.040562943228849e-05,
|
| 995 |
+
"loss": 0.2742,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5071428571428571,
|
| 1000 |
+
"grad_norm": 0.8784185700324445,
|
| 1001 |
+
"learning_rate": 1.0289774251630602e-05,
|
| 1002 |
+
"loss": 0.1844,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5107142857142857,
|
| 1007 |
+
"grad_norm": 1.105225096168923,
|
| 1008 |
+
"learning_rate": 1.0173880129166358e-05,
|
| 1009 |
+
"loss": 0.2664,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5142857142857142,
|
| 1014 |
+
"grad_norm": 0.7220119823796761,
|
| 1015 |
+
"learning_rate": 1.0057962639524799e-05,
|
| 1016 |
+
"loss": 0.1911,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5178571428571429,
|
| 1021 |
+
"grad_norm": 0.9901039928611911,
|
| 1022 |
+
"learning_rate": 9.942037360475205e-06,
|
| 1023 |
+
"loss": 0.2047,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5214285714285715,
|
| 1028 |
+
"grad_norm": 0.990343263513173,
|
| 1029 |
+
"learning_rate": 9.826119870833644e-06,
|
| 1030 |
+
"loss": 0.2482,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.525,
|
| 1035 |
+
"grad_norm": 1.0814804344690816,
|
| 1036 |
+
"learning_rate": 9.710225748369402e-06,
|
| 1037 |
+
"loss": 0.2791,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5285714285714286,
|
| 1042 |
+
"grad_norm": 0.6699567812276747,
|
| 1043 |
+
"learning_rate": 9.594370567711512e-06,
|
| 1044 |
+
"loss": 0.1689,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5321428571428571,
|
| 1049 |
+
"grad_norm": 1.193514972349772,
|
| 1050 |
+
"learning_rate": 9.478569898255765e-06,
|
| 1051 |
+
"loss": 0.3267,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5357142857142857,
|
| 1056 |
+
"grad_norm": 0.718259276227398,
|
| 1057 |
+
"learning_rate": 9.362839302072354e-06,
|
| 1058 |
+
"loss": 0.133,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5392857142857143,
|
| 1063 |
+
"grad_norm": 1.071572674237938,
|
| 1064 |
+
"learning_rate": 9.247194331814561e-06,
|
| 1065 |
+
"loss": 0.2434,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5428571428571428,
|
| 1070 |
+
"grad_norm": 1.0026565147511441,
|
| 1071 |
+
"learning_rate": 9.131650528628688e-06,
|
| 1072 |
+
"loss": 0.2248,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5464285714285714,
|
| 1077 |
+
"grad_norm": 0.7606699824597489,
|
| 1078 |
+
"learning_rate": 9.016223420065519e-06,
|
| 1079 |
+
"loss": 0.2294,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.55,
|
| 1084 |
+
"grad_norm": 0.9509831030302495,
|
| 1085 |
+
"learning_rate": 8.900928517993644e-06,
|
| 1086 |
+
"loss": 0.2921,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5535714285714286,
|
| 1091 |
+
"grad_norm": 0.9871198309746722,
|
| 1092 |
+
"learning_rate": 8.785781316514841e-06,
|
| 1093 |
+
"loss": 0.2715,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5571428571428572,
|
| 1098 |
+
"grad_norm": 0.9057094151212118,
|
| 1099 |
+
"learning_rate": 8.670797289881915e-06,
|
| 1100 |
+
"loss": 0.2449,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5607142857142857,
|
| 1105 |
+
"grad_norm": 0.8772737026647174,
|
| 1106 |
+
"learning_rate": 8.555991890419116e-06,
|
| 1107 |
+
"loss": 0.2564,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5642857142857143,
|
| 1112 |
+
"grad_norm": 0.8695626353738833,
|
| 1113 |
+
"learning_rate": 8.441380546445603e-06,
|
| 1114 |
+
"loss": 0.2229,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5678571428571428,
|
| 1119 |
+
"grad_norm": 1.073152464539624,
|
| 1120 |
+
"learning_rate": 8.326978660202034e-06,
|
| 1121 |
+
"loss": 0.3386,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5714285714285714,
|
| 1126 |
+
"grad_norm": 1.012416344543051,
|
| 1127 |
+
"learning_rate": 8.212801605780754e-06,
|
| 1128 |
+
"loss": 0.2207,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.575,
|
| 1133 |
+
"grad_norm": 0.7691240257824403,
|
| 1134 |
+
"learning_rate": 8.098864727059685e-06,
|
| 1135 |
+
"loss": 0.2051,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5785714285714286,
|
| 1140 |
+
"grad_norm": 1.1387099994481609,
|
| 1141 |
+
"learning_rate": 7.985183335640332e-06,
|
| 1142 |
+
"loss": 0.3062,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5821428571428572,
|
| 1147 |
+
"grad_norm": 0.6548715665719999,
|
| 1148 |
+
"learning_rate": 7.871772708790114e-06,
|
| 1149 |
+
"loss": 0.1671,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5857142857142857,
|
| 1154 |
+
"grad_norm": 1.0074068488831938,
|
| 1155 |
+
"learning_rate": 7.758648087389277e-06,
|
| 1156 |
+
"loss": 0.2457,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5892857142857143,
|
| 1161 |
+
"grad_norm": 1.3590905269175524,
|
| 1162 |
+
"learning_rate": 7.64582467388275e-06,
|
| 1163 |
+
"loss": 0.2763,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5928571428571429,
|
| 1168 |
+
"grad_norm": 0.6688733178759887,
|
| 1169 |
+
"learning_rate": 7.533317630237117e-06,
|
| 1170 |
+
"loss": 0.2107,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5964285714285714,
|
| 1175 |
+
"grad_norm": 1.1521044685448598,
|
| 1176 |
+
"learning_rate": 7.421142075903067e-06,
|
| 1177 |
+
"loss": 0.3182,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6,
|
| 1182 |
+
"grad_norm": 0.8663913856359603,
|
| 1183 |
+
"learning_rate": 7.3093130857835245e-06,
|
| 1184 |
+
"loss": 0.1907,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6035714285714285,
|
| 1189 |
+
"grad_norm": 1.0039720576945341,
|
| 1190 |
+
"learning_rate": 7.197845688207805e-06,
|
| 1191 |
+
"loss": 0.2518,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6071428571428571,
|
| 1196 |
+
"grad_norm": 0.8382491705573645,
|
| 1197 |
+
"learning_rate": 7.086754862911982e-06,
|
| 1198 |
+
"loss": 0.2304,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6107142857142858,
|
| 1203 |
+
"grad_norm": 1.102459135393568,
|
| 1204 |
+
"learning_rate": 6.976055539025819e-06,
|
| 1205 |
+
"loss": 0.2607,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6142857142857143,
|
| 1210 |
+
"grad_norm": 0.8659998101978996,
|
| 1211 |
+
"learning_rate": 6.865762593066514e-06,
|
| 1212 |
+
"loss": 0.2173,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6178571428571429,
|
| 1217 |
+
"grad_norm": 1.038371053979899,
|
| 1218 |
+
"learning_rate": 6.755890846939454e-06,
|
| 1219 |
+
"loss": 0.2449,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6214285714285714,
|
| 1224 |
+
"grad_norm": 0.9112242172551936,
|
| 1225 |
+
"learning_rate": 6.646455065946386e-06,
|
| 1226 |
+
"loss": 0.2446,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.625,
|
| 1231 |
+
"grad_norm": 0.8416570702077658,
|
| 1232 |
+
"learning_rate": 6.537469956801128e-06,
|
| 1233 |
+
"loss": 0.2074,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6285714285714286,
|
| 1238 |
+
"grad_norm": 0.7222497910641669,
|
| 1239 |
+
"learning_rate": 6.428950165653204e-06,
|
| 1240 |
+
"loss": 0.2037,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6321428571428571,
|
| 1245 |
+
"grad_norm": 0.9350757855602984,
|
| 1246 |
+
"learning_rate": 6.320910276119576e-06,
|
| 1247 |
+
"loss": 0.26,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6357142857142857,
|
| 1252 |
+
"grad_norm": 0.8975477713777471,
|
| 1253 |
+
"learning_rate": 6.213364807324817e-06,
|
| 1254 |
+
"loss": 0.2336,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6392857142857142,
|
| 1259 |
+
"grad_norm": 0.713247964590351,
|
| 1260 |
+
"learning_rate": 6.106328211949928e-06,
|
| 1261 |
+
"loss": 0.1881,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6428571428571429,
|
| 1266 |
+
"grad_norm": 0.7432676487695248,
|
| 1267 |
+
"learning_rate": 5.999814874290084e-06,
|
| 1268 |
+
"loss": 0.188,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6464285714285715,
|
| 1273 |
+
"grad_norm": 0.9375433226321466,
|
| 1274 |
+
"learning_rate": 5.893839108321584e-06,
|
| 1275 |
+
"loss": 0.2229,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.65,
|
| 1280 |
+
"grad_norm": 0.9035178928866511,
|
| 1281 |
+
"learning_rate": 5.7884151557782305e-06,
|
| 1282 |
+
"loss": 0.268,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6535714285714286,
|
| 1287 |
+
"grad_norm": 1.0861928031153827,
|
| 1288 |
+
"learning_rate": 5.68355718423746e-06,
|
| 1289 |
+
"loss": 0.2536,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6571428571428571,
|
| 1294 |
+
"grad_norm": 0.9249874842289382,
|
| 1295 |
+
"learning_rate": 5.579279285216369e-06,
|
| 1296 |
+
"loss": 0.2984,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6607142857142857,
|
| 1301 |
+
"grad_norm": 0.9641440586255716,
|
| 1302 |
+
"learning_rate": 5.4755954722780236e-06,
|
| 1303 |
+
"loss": 0.2843,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6642857142857143,
|
| 1308 |
+
"grad_norm": 1.047518256416677,
|
| 1309 |
+
"learning_rate": 5.372519679148227e-06,
|
| 1310 |
+
"loss": 0.2846,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6678571428571428,
|
| 1315 |
+
"grad_norm": 0.8899779907249936,
|
| 1316 |
+
"learning_rate": 5.270065757843e-06,
|
| 1317 |
+
"loss": 0.2814,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6714285714285714,
|
| 1322 |
+
"grad_norm": 1.257826620638329,
|
| 1323 |
+
"learning_rate": 5.168247476807054e-06,
|
| 1324 |
+
"loss": 0.3616,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.675,
|
| 1329 |
+
"grad_norm": 0.8378870170037112,
|
| 1330 |
+
"learning_rate": 5.067078519063514e-06,
|
| 1331 |
+
"loss": 0.206,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6785714285714286,
|
| 1336 |
+
"grad_norm": 0.8499116136055077,
|
| 1337 |
+
"learning_rate": 4.966572480375076e-06,
|
| 1338 |
+
"loss": 0.2655,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6821428571428572,
|
| 1343 |
+
"grad_norm": 1.0500985457733107,
|
| 1344 |
+
"learning_rate": 4.86674286741693e-06,
|
| 1345 |
+
"loss": 0.2423,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6857142857142857,
|
| 1350 |
+
"grad_norm": 1.1824577867055674,
|
| 1351 |
+
"learning_rate": 4.767603095961652e-06,
|
| 1352 |
+
"loss": 0.3178,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6892857142857143,
|
| 1357 |
+
"grad_norm": 1.0135768392531344,
|
| 1358 |
+
"learning_rate": 4.669166489076283e-06,
|
| 1359 |
+
"loss": 0.3272,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6928571428571428,
|
| 1364 |
+
"grad_norm": 0.7599653832936717,
|
| 1365 |
+
"learning_rate": 4.571446275331903e-06,
|
| 1366 |
+
"loss": 0.2039,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6964285714285714,
|
| 1371 |
+
"grad_norm": 1.0212451468927062,
|
| 1372 |
+
"learning_rate": 4.47445558702587e-06,
|
| 1373 |
+
"loss": 0.285,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7,
|
| 1378 |
+
"grad_norm": 0.9471963980854589,
|
| 1379 |
+
"learning_rate": 4.378207458417035e-06,
|
| 1380 |
+
"loss": 0.2724,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7035714285714286,
|
| 1385 |
+
"grad_norm": 0.8971393901363991,
|
| 1386 |
+
"learning_rate": 4.282714823974088e-06,
|
| 1387 |
+
"loss": 0.1893,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.7071428571428572,
|
| 1392 |
+
"grad_norm": 0.9659760236519076,
|
| 1393 |
+
"learning_rate": 4.187990516637361e-06,
|
| 1394 |
+
"loss": 0.2423,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7107142857142857,
|
| 1399 |
+
"grad_norm": 0.8988943829563382,
|
| 1400 |
+
"learning_rate": 4.094047266094225e-06,
|
| 1401 |
+
"loss": 0.2479,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7142857142857143,
|
| 1406 |
+
"grad_norm": 0.6167861266259378,
|
| 1407 |
+
"learning_rate": 4.000897697068418e-06,
|
| 1408 |
+
"loss": 0.186,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7178571428571429,
|
| 1413 |
+
"grad_norm": 0.9931841997154047,
|
| 1414 |
+
"learning_rate": 3.908554327623425e-06,
|
| 1415 |
+
"loss": 0.2703,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7214285714285714,
|
| 1420 |
+
"grad_norm": 1.0515423629697425,
|
| 1421 |
+
"learning_rate": 3.817029567480228e-06,
|
| 1422 |
+
"loss": 0.2872,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.725,
|
| 1427 |
+
"grad_norm": 1.0078364499059038,
|
| 1428 |
+
"learning_rate": 3.7263357163496118e-06,
|
| 1429 |
+
"loss": 0.2771,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7285714285714285,
|
| 1434 |
+
"grad_norm": 0.7946329327315834,
|
| 1435 |
+
"learning_rate": 3.6364849622792262e-06,
|
| 1436 |
+
"loss": 0.2636,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.7321428571428571,
|
| 1441 |
+
"grad_norm": 1.0961952413792404,
|
| 1442 |
+
"learning_rate": 3.5474893800157005e-06,
|
| 1443 |
+
"loss": 0.2644,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7357142857142858,
|
| 1448 |
+
"grad_norm": 0.8310362218238262,
|
| 1449 |
+
"learning_rate": 3.459360929381931e-06,
|
| 1450 |
+
"loss": 0.1621,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.7392857142857143,
|
| 1455 |
+
"grad_norm": 0.9112572267088846,
|
| 1456 |
+
"learning_rate": 3.372111453669864e-06,
|
| 1457 |
+
"loss": 0.2583,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7428571428571429,
|
| 1462 |
+
"grad_norm": 0.8328762351773727,
|
| 1463 |
+
"learning_rate": 3.2857526780488925e-06,
|
| 1464 |
+
"loss": 0.252,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7464285714285714,
|
| 1469 |
+
"grad_norm": 0.7790083845874031,
|
| 1470 |
+
"learning_rate": 3.2002962079901743e-06,
|
| 1471 |
+
"loss": 0.1739,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.75,
|
| 1476 |
+
"grad_norm": 0.8937099611163849,
|
| 1477 |
+
"learning_rate": 3.115753527706986e-06,
|
| 1478 |
+
"loss": 0.2641,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7535714285714286,
|
| 1483 |
+
"grad_norm": 0.665235869076356,
|
| 1484 |
+
"learning_rate": 3.0321359986114096e-06,
|
| 1485 |
+
"loss": 0.1929,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7571428571428571,
|
| 1490 |
+
"grad_norm": 0.9975910293748517,
|
| 1491 |
+
"learning_rate": 2.9494548577875195e-06,
|
| 1492 |
+
"loss": 0.2071,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7607142857142857,
|
| 1497 |
+
"grad_norm": 0.8635252751329144,
|
| 1498 |
+
"learning_rate": 2.8677212164812464e-06,
|
| 1499 |
+
"loss": 0.2123,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7642857142857142,
|
| 1504 |
+
"grad_norm": 0.9914555320649461,
|
| 1505 |
+
"learning_rate": 2.786946058607187e-06,
|
| 1506 |
+
"loss": 0.3021,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7678571428571429,
|
| 1511 |
+
"grad_norm": 1.067677741156891,
|
| 1512 |
+
"learning_rate": 2.70714023927251e-06,
|
| 1513 |
+
"loss": 0.3197,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7714285714285715,
|
| 1518 |
+
"grad_norm": 1.0127321773951852,
|
| 1519 |
+
"learning_rate": 2.628314483318178e-06,
|
| 1520 |
+
"loss": 0.2979,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.775,
|
| 1525 |
+
"grad_norm": 0.8879989714883786,
|
| 1526 |
+
"learning_rate": 2.5504793838776585e-06,
|
| 1527 |
+
"loss": 0.2181,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7785714285714286,
|
| 1532 |
+
"grad_norm": 1.0116373979132076,
|
| 1533 |
+
"learning_rate": 2.473645400953366e-06,
|
| 1534 |
+
"loss": 0.3007,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7821428571428571,
|
| 1539 |
+
"grad_norm": 0.8314841547996328,
|
| 1540 |
+
"learning_rate": 2.3978228600109564e-06,
|
| 1541 |
+
"loss": 0.275,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7857142857142857,
|
| 1546 |
+
"grad_norm": 0.9162454344653038,
|
| 1547 |
+
"learning_rate": 2.323021950591743e-06,
|
| 1548 |
+
"loss": 0.2748,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7892857142857143,
|
| 1553 |
+
"grad_norm": 1.0145382352758672,
|
| 1554 |
+
"learning_rate": 2.249252724943336e-06,
|
| 1555 |
+
"loss": 0.2701,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7928571428571428,
|
| 1560 |
+
"grad_norm": 0.8666753235531185,
|
| 1561 |
+
"learning_rate": 2.176525096668769e-06,
|
| 1562 |
+
"loss": 0.254,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7964285714285714,
|
| 1567 |
+
"grad_norm": 0.8928533620858623,
|
| 1568 |
+
"learning_rate": 2.1048488393942455e-06,
|
| 1569 |
+
"loss": 0.2623,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.8,
|
| 1574 |
+
"grad_norm": 0.7977864862031556,
|
| 1575 |
+
"learning_rate": 2.0342335854556738e-06,
|
| 1576 |
+
"loss": 0.2331,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.8035714285714286,
|
| 1581 |
+
"grad_norm": 0.9396848764074734,
|
| 1582 |
+
"learning_rate": 1.964688824604234e-06,
|
| 1583 |
+
"loss": 0.2283,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8071428571428572,
|
| 1588 |
+
"grad_norm": 0.7043691856433903,
|
| 1589 |
+
"learning_rate": 1.896223902731058e-06,
|
| 1590 |
+
"loss": 0.1487,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8107142857142857,
|
| 1595 |
+
"grad_norm": 0.8028110829154207,
|
| 1596 |
+
"learning_rate": 1.8288480206112879e-06,
|
| 1597 |
+
"loss": 0.178,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8142857142857143,
|
| 1602 |
+
"grad_norm": 0.9551250104442663,
|
| 1603 |
+
"learning_rate": 1.7625702326675952e-06,
|
| 1604 |
+
"loss": 0.2512,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.8178571428571428,
|
| 1609 |
+
"grad_norm": 0.8542011521297516,
|
| 1610 |
+
"learning_rate": 1.6973994457534026e-06,
|
| 1611 |
+
"loss": 0.2858,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8214285714285714,
|
| 1616 |
+
"grad_norm": 0.8529278139532056,
|
| 1617 |
+
"learning_rate": 1.6333444179559078e-06,
|
| 1618 |
+
"loss": 0.2351,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.825,
|
| 1623 |
+
"grad_norm": 0.9391405907710427,
|
| 1624 |
+
"learning_rate": 1.5704137574191202e-06,
|
| 1625 |
+
"loss": 0.2612,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8285714285714286,
|
| 1630 |
+
"grad_norm": 0.9898771118350357,
|
| 1631 |
+
"learning_rate": 1.5086159211870445e-06,
|
| 1632 |
+
"loss": 0.2806,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.8321428571428572,
|
| 1637 |
+
"grad_norm": 1.3372111053303188,
|
| 1638 |
+
"learning_rate": 1.447959214067155e-06,
|
| 1639 |
+
"loss": 0.3463,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.8357142857142857,
|
| 1644 |
+
"grad_norm": 0.8321272930723775,
|
| 1645 |
+
"learning_rate": 1.3884517875143544e-06,
|
| 1646 |
+
"loss": 0.2537,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.8392857142857143,
|
| 1651 |
+
"grad_norm": 1.2134366545999935,
|
| 1652 |
+
"learning_rate": 1.3301016385355093e-06,
|
| 1653 |
+
"loss": 0.3719,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8428571428571429,
|
| 1658 |
+
"grad_norm": 0.6334539224525021,
|
| 1659 |
+
"learning_rate": 1.2729166086147803e-06,
|
| 1660 |
+
"loss": 0.1432,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8464285714285714,
|
| 1665 |
+
"grad_norm": 0.876155106277133,
|
| 1666 |
+
"learning_rate": 1.216904382659806e-06,
|
| 1667 |
+
"loss": 0.2743,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.85,
|
| 1672 |
+
"grad_norm": 1.3009308060492257,
|
| 1673 |
+
"learning_rate": 1.1620724879689793e-06,
|
| 1674 |
+
"loss": 0.3375,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8535714285714285,
|
| 1679 |
+
"grad_norm": 0.8890031161074461,
|
| 1680 |
+
"learning_rate": 1.1084282932198543e-06,
|
| 1681 |
+
"loss": 0.2675,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8571428571428571,
|
| 1686 |
+
"grad_norm": 0.9590774137550018,
|
| 1687 |
+
"learning_rate": 1.0559790074789134e-06,
|
| 1688 |
+
"loss": 0.2847,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8607142857142858,
|
| 1693 |
+
"grad_norm": 0.9091813907584589,
|
| 1694 |
+
"learning_rate": 1.00473167923275e-06,
|
| 1695 |
+
"loss": 0.2142,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8642857142857143,
|
| 1700 |
+
"grad_norm": 1.1904931385168855,
|
| 1701 |
+
"learning_rate": 9.546931954408622e-07,
|
| 1702 |
+
"loss": 0.2657,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8678571428571429,
|
| 1707 |
+
"grad_norm": 1.0331566604474673,
|
| 1708 |
+
"learning_rate": 9.058702806101172e-07,
|
| 1709 |
+
"loss": 0.2839,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8714285714285714,
|
| 1714 |
+
"grad_norm": 0.8801867878946903,
|
| 1715 |
+
"learning_rate": 8.582694958910809e-07,
|
| 1716 |
+
"loss": 0.2513,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.875,
|
| 1721 |
+
"grad_norm": 1.2746454104318332,
|
| 1722 |
+
"learning_rate": 8.118972381962853e-07,
|
| 1723 |
+
"loss": 0.475,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8785714285714286,
|
| 1728 |
+
"grad_norm": 0.8860673740110999,
|
| 1729 |
+
"learning_rate": 7.667597393405602e-07,
|
| 1730 |
+
"loss": 0.2575,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8821428571428571,
|
| 1735 |
+
"grad_norm": 0.7619213943421504,
|
| 1736 |
+
"learning_rate": 7.228630652035717e-07,
|
| 1737 |
+
"loss": 0.2073,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8857142857142857,
|
| 1742 |
+
"grad_norm": 1.024455123403684,
|
| 1743 |
+
"learning_rate": 6.802131149146374e-07,
|
| 1744 |
+
"loss": 0.2807,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8892857142857142,
|
| 1749 |
+
"grad_norm": 0.8835926745414224,
|
| 1750 |
+
"learning_rate": 6.388156200599726e-07,
|
| 1751 |
+
"loss": 0.2636,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8928571428571429,
|
| 1756 |
+
"grad_norm": 0.7925007490357969,
|
| 1757 |
+
"learning_rate": 5.986761439124289e-07,
|
| 1758 |
+
"loss": 0.2308,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8964285714285715,
|
| 1763 |
+
"grad_norm": 1.3499511845536636,
|
| 1764 |
+
"learning_rate": 5.598000806838766e-07,
|
| 1765 |
+
"loss": 0.3801,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.9,
|
| 1770 |
+
"grad_norm": 0.7300222807825056,
|
| 1771 |
+
"learning_rate": 5.221926548002876e-07,
|
| 1772 |
+
"loss": 0.2236,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.9035714285714286,
|
| 1777 |
+
"grad_norm": 0.7658414820534243,
|
| 1778 |
+
"learning_rate": 4.858589201996433e-07,
|
| 1779 |
+
"loss": 0.2408,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9071428571428571,
|
| 1784 |
+
"grad_norm": 0.8451965731847849,
|
| 1785 |
+
"learning_rate": 4.5080375965275256e-07,
|
| 1786 |
+
"loss": 0.241,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9107142857142857,
|
| 1791 |
+
"grad_norm": 0.9763077207541743,
|
| 1792 |
+
"learning_rate": 4.1703188410707087e-07,
|
| 1793 |
+
"loss": 0.2934,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.9142857142857143,
|
| 1798 |
+
"grad_norm": 1.0450416449938653,
|
| 1799 |
+
"learning_rate": 3.845478320536178e-07,
|
| 1800 |
+
"loss": 0.3037,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.9178571428571428,
|
| 1805 |
+
"grad_norm": 1.0924428398328787,
|
| 1806 |
+
"learning_rate": 3.5335596891705406e-07,
|
| 1807 |
+
"loss": 0.2839,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.9214285714285714,
|
| 1812 |
+
"grad_norm": 0.7920945565609422,
|
| 1813 |
+
"learning_rate": 3.2346048646903494e-07,
|
| 1814 |
+
"loss": 0.2368,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.925,
|
| 1819 |
+
"grad_norm": 0.9915364248763068,
|
| 1820 |
+
"learning_rate": 2.9486540226488556e-07,
|
| 1821 |
+
"loss": 0.2638,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9285714285714286,
|
| 1826 |
+
"grad_norm": 0.9701488675520441,
|
| 1827 |
+
"learning_rate": 2.6757455910370486e-07,
|
| 1828 |
+
"loss": 0.2348,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.9321428571428572,
|
| 1833 |
+
"grad_norm": 0.7205020345952264,
|
| 1834 |
+
"learning_rate": 2.41591624511931e-07,
|
| 1835 |
+
"loss": 0.2275,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.9357142857142857,
|
| 1840 |
+
"grad_norm": 0.9145542664276822,
|
| 1841 |
+
"learning_rate": 2.169200902504842e-07,
|
| 1842 |
+
"loss": 0.2504,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.9392857142857143,
|
| 1847 |
+
"grad_norm": 0.9366813037254895,
|
| 1848 |
+
"learning_rate": 1.9356327184551716e-07,
|
| 1849 |
+
"loss": 0.2359,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.9428571428571428,
|
| 1854 |
+
"grad_norm": 0.7940417127042732,
|
| 1855 |
+
"learning_rate": 1.7152430814285303e-07,
|
| 1856 |
+
"loss": 0.2657,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9464285714285714,
|
| 1861 |
+
"grad_norm": 1.0837046668744013,
|
| 1862 |
+
"learning_rate": 1.5080616088616884e-07,
|
| 1863 |
+
"loss": 0.2948,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.95,
|
| 1868 |
+
"grad_norm": 0.9318179117611772,
|
| 1869 |
+
"learning_rate": 1.3141161431896809e-07,
|
| 1870 |
+
"loss": 0.3247,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9535714285714286,
|
| 1875 |
+
"grad_norm": 0.8452839653435972,
|
| 1876 |
+
"learning_rate": 1.1334327481042573e-07,
|
| 1877 |
+
"loss": 0.185,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.9571428571428572,
|
| 1882 |
+
"grad_norm": 0.9398904285158833,
|
| 1883 |
+
"learning_rate": 9.660357050512158e-08,
|
| 1884 |
+
"loss": 0.2707,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9607142857142857,
|
| 1889 |
+
"grad_norm": 1.242666900837473,
|
| 1890 |
+
"learning_rate": 8.119475099673035e-08,
|
| 1891 |
+
"loss": 0.3251,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9642857142857143,
|
| 1896 |
+
"grad_norm": 0.788276270917447,
|
| 1897 |
+
"learning_rate": 6.711888702570556e-08,
|
| 1898 |
+
"loss": 0.2321,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9678571428571429,
|
| 1903 |
+
"grad_norm": 0.8206210297411658,
|
| 1904 |
+
"learning_rate": 5.437787020100116e-08,
|
| 1905 |
+
"loss": 0.2181,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9714285714285714,
|
| 1910 |
+
"grad_norm": 1.0131859227874498,
|
| 1911 |
+
"learning_rate": 4.297341274586475e-08,
|
| 1912 |
+
"loss": 0.3155,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.975,
|
| 1917 |
+
"grad_norm": 1.0890650915672193,
|
| 1918 |
+
"learning_rate": 3.290704726773619e-08,
|
| 1919 |
+
"loss": 0.2678,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9785714285714285,
|
| 1924 |
+
"grad_norm": 1.0442519996924977,
|
| 1925 |
+
"learning_rate": 2.4180126552284523e-08,
|
| 1926 |
+
"loss": 0.2663,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.9821428571428571,
|
| 1931 |
+
"grad_norm": 1.0298701388046845,
|
| 1932 |
+
"learning_rate": 1.6793823381614506e-08,
|
| 1933 |
+
"loss": 0.2908,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.9857142857142858,
|
| 1938 |
+
"grad_norm": 0.9020170998081513,
|
| 1939 |
+
"learning_rate": 1.0749130376659366e-08,
|
| 1940 |
+
"loss": 0.2456,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9892857142857143,
|
| 1945 |
+
"grad_norm": 0.9361316642559707,
|
| 1946 |
+
"learning_rate": 6.046859863781951e-09,
|
| 1947 |
+
"loss": 0.3191,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9928571428571429,
|
| 1952 |
+
"grad_norm": 0.9664382332584418,
|
| 1953 |
+
"learning_rate": 2.6876437656153663e-09,
|
| 1954 |
+
"loss": 0.2245,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.9964285714285714,
|
| 1959 |
+
"grad_norm": 0.8441442143857916,
|
| 1960 |
+
"learning_rate": 6.719335161364804e-10,
|
| 1961 |
+
"loss": 0.2086,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 1.0,
|
| 1966 |
+
"grad_norm": 1.085337268493026,
|
| 1967 |
+
"learning_rate": 0.0,
|
| 1968 |
+
"loss": 0.2965,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 1.0,
|
| 1973 |
+
"step": 280,
|
| 1974 |
+
"total_flos": 146328534351872.0,
|
| 1975 |
+
"train_loss": 0.4051302823637213,
|
| 1976 |
+
"train_runtime": 1179.9378,
|
| 1977 |
+
"train_samples_per_second": 30.364,
|
| 1978 |
+
"train_steps_per_second": 0.237
|
| 1979 |
+
}
|
| 1980 |
+
],
|
| 1981 |
+
"logging_steps": 1.0,
|
| 1982 |
+
"max_steps": 280,
|
| 1983 |
+
"num_input_tokens_seen": 0,
|
| 1984 |
+
"num_train_epochs": 1,
|
| 1985 |
+
"save_steps": 50000,
|
| 1986 |
+
"stateful_callbacks": {
|
| 1987 |
+
"TrainerControl": {
|
| 1988 |
+
"args": {
|
| 1989 |
+
"should_epoch_stop": false,
|
| 1990 |
+
"should_evaluate": false,
|
| 1991 |
+
"should_log": false,
|
| 1992 |
+
"should_save": false,
|
| 1993 |
+
"should_training_stop": false
|
| 1994 |
+
},
|
| 1995 |
+
"attributes": {}
|
| 1996 |
+
}
|
| 1997 |
+
},
|
| 1998 |
+
"total_flos": 146328534351872.0,
|
| 1999 |
+
"train_batch_size": 4,
|
| 2000 |
+
"trial_name": null,
|
| 2001 |
+
"trial_params": null
|
| 2002 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94fdcdad5331c1fe7cf3c0522ceffb7eb233486961402861304c5296d698762c
|
| 3 |
+
size 7224
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|