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-Qwen2-1.5B-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": 1536,
|
| 12 |
+
"image_aspect_ratio": "resizesquare",
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8960,
|
| 15 |
+
"max_position_embeddings": 32768,
|
| 16 |
+
"max_window_layers": 28,
|
| 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": 12,
|
| 28 |
+
"num_hidden_layers": 28,
|
| 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 sha256:0b9f9b4da6203af90f197a03edad376fbfd507bfe65ca140dda9a0a075965efe
|
| 3 |
+
size 3702414320
|
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": 3.2543029662772347,
|
| 14 |
+
"learning_rate": 2.222222222222222e-06,
|
| 15 |
+
"loss": 1.554,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007142857142857143,
|
| 20 |
+
"grad_norm": 2.7624244157844706,
|
| 21 |
+
"learning_rate": 4.444444444444444e-06,
|
| 22 |
+
"loss": 1.4127,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010714285714285714,
|
| 27 |
+
"grad_norm": 3.1002251906462344,
|
| 28 |
+
"learning_rate": 6.666666666666667e-06,
|
| 29 |
+
"loss": 1.4643,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014285714285714285,
|
| 34 |
+
"grad_norm": 2.3755712686804804,
|
| 35 |
+
"learning_rate": 8.888888888888888e-06,
|
| 36 |
+
"loss": 1.3791,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.017857142857142856,
|
| 41 |
+
"grad_norm": 1.9308377348791768,
|
| 42 |
+
"learning_rate": 1.1111111111111113e-05,
|
| 43 |
+
"loss": 1.3221,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02142857142857143,
|
| 48 |
+
"grad_norm": 2.0118553962006427,
|
| 49 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 50 |
+
"loss": 1.2667,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.025,
|
| 55 |
+
"grad_norm": 2.249667975836266,
|
| 56 |
+
"learning_rate": 1.555555555555556e-05,
|
| 57 |
+
"loss": 1.3315,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02857142857142857,
|
| 62 |
+
"grad_norm": 3.3286044973535236,
|
| 63 |
+
"learning_rate": 1.7777777777777777e-05,
|
| 64 |
+
"loss": 1.2336,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03214285714285714,
|
| 69 |
+
"grad_norm": 2.607483213778446,
|
| 70 |
+
"learning_rate": 2e-05,
|
| 71 |
+
"loss": 1.2549,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03571428571428571,
|
| 76 |
+
"grad_norm": 1.9059436966580536,
|
| 77 |
+
"learning_rate": 1.9999328066483867e-05,
|
| 78 |
+
"loss": 1.1484,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.039285714285714285,
|
| 83 |
+
"grad_norm": 2.351052203479045,
|
| 84 |
+
"learning_rate": 1.9997312356234385e-05,
|
| 85 |
+
"loss": 1.1006,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04285714285714286,
|
| 90 |
+
"grad_norm": 2.287439785470941,
|
| 91 |
+
"learning_rate": 1.999395314013622e-05,
|
| 92 |
+
"loss": 1.0522,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04642857142857143,
|
| 97 |
+
"grad_norm": 1.7697435082139328,
|
| 98 |
+
"learning_rate": 1.998925086962334e-05,
|
| 99 |
+
"loss": 1.0117,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05,
|
| 104 |
+
"grad_norm": 1.7809725059300023,
|
| 105 |
+
"learning_rate": 1.998320617661839e-05,
|
| 106 |
+
"loss": 0.9512,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05357142857142857,
|
| 111 |
+
"grad_norm": 1.6904647102656667,
|
| 112 |
+
"learning_rate": 1.997581987344772e-05,
|
| 113 |
+
"loss": 0.9306,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05714285714285714,
|
| 118 |
+
"grad_norm": 1.6895690138180741,
|
| 119 |
+
"learning_rate": 1.9967092952732266e-05,
|
| 120 |
+
"loss": 0.8694,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.060714285714285714,
|
| 125 |
+
"grad_norm": 1.7669079061201738,
|
| 126 |
+
"learning_rate": 1.9957026587254136e-05,
|
| 127 |
+
"loss": 0.8984,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06428571428571428,
|
| 132 |
+
"grad_norm": 1.476756594220327,
|
| 133 |
+
"learning_rate": 1.9945622129799e-05,
|
| 134 |
+
"loss": 0.8041,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06785714285714285,
|
| 139 |
+
"grad_norm": 1.5474239491243957,
|
| 140 |
+
"learning_rate": 1.9932881112974298e-05,
|
| 141 |
+
"loss": 0.8109,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07142857142857142,
|
| 146 |
+
"grad_norm": 1.449807734275577,
|
| 147 |
+
"learning_rate": 1.9918805249003272e-05,
|
| 148 |
+
"loss": 0.6632,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.075,
|
| 153 |
+
"grad_norm": 1.4334867898446597,
|
| 154 |
+
"learning_rate": 1.9903396429494882e-05,
|
| 155 |
+
"loss": 0.6602,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07857142857142857,
|
| 160 |
+
"grad_norm": 1.3803101146379801,
|
| 161 |
+
"learning_rate": 1.9886656725189575e-05,
|
| 162 |
+
"loss": 0.6428,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08214285714285714,
|
| 167 |
+
"grad_norm": 1.3124607153366843,
|
| 168 |
+
"learning_rate": 1.9868588385681035e-05,
|
| 169 |
+
"loss": 0.6175,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08571428571428572,
|
| 174 |
+
"grad_norm": 1.23178251089412,
|
| 175 |
+
"learning_rate": 1.9849193839113833e-05,
|
| 176 |
+
"loss": 0.5117,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08928571428571429,
|
| 181 |
+
"grad_norm": 1.1688906258052285,
|
| 182 |
+
"learning_rate": 1.9828475691857148e-05,
|
| 183 |
+
"loss": 0.5335,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09285714285714286,
|
| 188 |
+
"grad_norm": 1.1144465197033029,
|
| 189 |
+
"learning_rate": 1.9806436728154484e-05,
|
| 190 |
+
"loss": 0.5583,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.09642857142857143,
|
| 195 |
+
"grad_norm": 1.2143115091886203,
|
| 196 |
+
"learning_rate": 1.9783079909749516e-05,
|
| 197 |
+
"loss": 0.5482,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.1,
|
| 202 |
+
"grad_norm": 1.2235901463474592,
|
| 203 |
+
"learning_rate": 1.975840837548807e-05,
|
| 204 |
+
"loss": 0.4723,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10357142857142858,
|
| 209 |
+
"grad_norm": 1.1827323503795928,
|
| 210 |
+
"learning_rate": 1.9732425440896298e-05,
|
| 211 |
+
"loss": 0.5831,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.10714285714285714,
|
| 216 |
+
"grad_norm": 1.1949009678747529,
|
| 217 |
+
"learning_rate": 1.9705134597735113e-05,
|
| 218 |
+
"loss": 0.4932,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11071428571428571,
|
| 223 |
+
"grad_norm": 1.1096843684141346,
|
| 224 |
+
"learning_rate": 1.9676539513530967e-05,
|
| 225 |
+
"loss": 0.3598,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11428571428571428,
|
| 230 |
+
"grad_norm": 1.0706468088750498,
|
| 231 |
+
"learning_rate": 1.9646644031082948e-05,
|
| 232 |
+
"loss": 0.4306,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.11785714285714285,
|
| 237 |
+
"grad_norm": 1.5246249464687178,
|
| 238 |
+
"learning_rate": 1.9615452167946383e-05,
|
| 239 |
+
"loss": 0.5073,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12142857142857143,
|
| 244 |
+
"grad_norm": 1.2556676622859617,
|
| 245 |
+
"learning_rate": 1.958296811589293e-05,
|
| 246 |
+
"loss": 0.4762,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.125,
|
| 251 |
+
"grad_norm": 1.2909831452240883,
|
| 252 |
+
"learning_rate": 1.954919624034725e-05,
|
| 253 |
+
"loss": 0.4601,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.12857142857142856,
|
| 258 |
+
"grad_norm": 1.2644852139785117,
|
| 259 |
+
"learning_rate": 1.951414107980036e-05,
|
| 260 |
+
"loss": 0.3924,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13214285714285715,
|
| 265 |
+
"grad_norm": 1.177541703534973,
|
| 266 |
+
"learning_rate": 1.9477807345199717e-05,
|
| 267 |
+
"loss": 0.4298,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1357142857142857,
|
| 272 |
+
"grad_norm": 1.0788971626418669,
|
| 273 |
+
"learning_rate": 1.9440199919316125e-05,
|
| 274 |
+
"loss": 0.3866,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1392857142857143,
|
| 279 |
+
"grad_norm": 1.198940400239735,
|
| 280 |
+
"learning_rate": 1.9401323856087573e-05,
|
| 281 |
+
"loss": 0.4065,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14285714285714285,
|
| 286 |
+
"grad_norm": 1.120532645081093,
|
| 287 |
+
"learning_rate": 1.936118437994003e-05,
|
| 288 |
+
"loss": 0.4159,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.14642857142857144,
|
| 293 |
+
"grad_norm": 1.189893899476895,
|
| 294 |
+
"learning_rate": 1.9319786885085366e-05,
|
| 295 |
+
"loss": 0.3384,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15,
|
| 300 |
+
"grad_norm": 1.113585244588961,
|
| 301 |
+
"learning_rate": 1.927713693479643e-05,
|
| 302 |
+
"loss": 0.4105,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15357142857142858,
|
| 307 |
+
"grad_norm": 1.0519634925640304,
|
| 308 |
+
"learning_rate": 1.923324026065944e-05,
|
| 309 |
+
"loss": 0.3938,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.15714285714285714,
|
| 314 |
+
"grad_norm": 1.0101223695618087,
|
| 315 |
+
"learning_rate": 1.918810276180372e-05,
|
| 316 |
+
"loss": 0.352,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16071428571428573,
|
| 321 |
+
"grad_norm": 1.024763914116021,
|
| 322 |
+
"learning_rate": 1.9141730504108923e-05,
|
| 323 |
+
"loss": 0.3333,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16428571428571428,
|
| 328 |
+
"grad_norm": 0.9967381517754978,
|
| 329 |
+
"learning_rate": 1.9094129719389886e-05,
|
| 330 |
+
"loss": 0.3412,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.16785714285714284,
|
| 335 |
+
"grad_norm": 1.2637178696514346,
|
| 336 |
+
"learning_rate": 1.904530680455914e-05,
|
| 337 |
+
"loss": 0.3622,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17142857142857143,
|
| 342 |
+
"grad_norm": 1.0266452708065086,
|
| 343 |
+
"learning_rate": 1.8995268320767254e-05,
|
| 344 |
+
"loss": 0.3541,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.175,
|
| 349 |
+
"grad_norm": 1.0740239915454661,
|
| 350 |
+
"learning_rate": 1.894402099252109e-05,
|
| 351 |
+
"loss": 0.2964,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.17857142857142858,
|
| 356 |
+
"grad_norm": 1.069730824561292,
|
| 357 |
+
"learning_rate": 1.889157170678015e-05,
|
| 358 |
+
"loss": 0.3421,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18214285714285713,
|
| 363 |
+
"grad_norm": 0.9643772737738865,
|
| 364 |
+
"learning_rate": 1.8837927512031022e-05,
|
| 365 |
+
"loss": 0.3057,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.18571428571428572,
|
| 370 |
+
"grad_norm": 1.0652997055685356,
|
| 371 |
+
"learning_rate": 1.8783095617340193e-05,
|
| 372 |
+
"loss": 0.3,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.18928571428571428,
|
| 377 |
+
"grad_norm": 0.8166401050438583,
|
| 378 |
+
"learning_rate": 1.872708339138522e-05,
|
| 379 |
+
"loss": 0.2458,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19285714285714287,
|
| 384 |
+
"grad_norm": 0.8876770564223524,
|
| 385 |
+
"learning_rate": 1.866989836146449e-05,
|
| 386 |
+
"loss": 0.2515,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.19642857142857142,
|
| 391 |
+
"grad_norm": 1.003831099607791,
|
| 392 |
+
"learning_rate": 1.861154821248565e-05,
|
| 393 |
+
"loss": 0.32,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2,
|
| 398 |
+
"grad_norm": 0.9005609017037866,
|
| 399 |
+
"learning_rate": 1.8552040785932846e-05,
|
| 400 |
+
"loss": 0.3109,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.20357142857142857,
|
| 405 |
+
"grad_norm": 0.9196347354093852,
|
| 406 |
+
"learning_rate": 1.849138407881296e-05,
|
| 407 |
+
"loss": 0.2715,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.20714285714285716,
|
| 412 |
+
"grad_norm": 1.1659775560675805,
|
| 413 |
+
"learning_rate": 1.8429586242580884e-05,
|
| 414 |
+
"loss": 0.3161,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.21071428571428572,
|
| 419 |
+
"grad_norm": 0.8229225999880141,
|
| 420 |
+
"learning_rate": 1.8366655582044096e-05,
|
| 421 |
+
"loss": 0.1971,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21428571428571427,
|
| 426 |
+
"grad_norm": 1.0837498121968563,
|
| 427 |
+
"learning_rate": 1.83026005542466e-05,
|
| 428 |
+
"loss": 0.3534,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.21785714285714286,
|
| 433 |
+
"grad_norm": 0.9894720548109736,
|
| 434 |
+
"learning_rate": 1.8237429767332407e-05,
|
| 435 |
+
"loss": 0.2613,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22142857142857142,
|
| 440 |
+
"grad_norm": 0.9165029939399432,
|
| 441 |
+
"learning_rate": 1.8171151979388715e-05,
|
| 442 |
+
"loss": 0.232,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.225,
|
| 447 |
+
"grad_norm": 0.8903286251798155,
|
| 448 |
+
"learning_rate": 1.8103776097268942e-05,
|
| 449 |
+
"loss": 0.2857,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.22857142857142856,
|
| 454 |
+
"grad_norm": 0.8053044878739753,
|
| 455 |
+
"learning_rate": 1.803531117539577e-05,
|
| 456 |
+
"loss": 0.2177,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.23214285714285715,
|
| 461 |
+
"grad_norm": 0.8248138749411966,
|
| 462 |
+
"learning_rate": 1.7965766414544328e-05,
|
| 463 |
+
"loss": 0.1982,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2357142857142857,
|
| 468 |
+
"grad_norm": 0.9377249996982162,
|
| 469 |
+
"learning_rate": 1.7895151160605758e-05,
|
| 470 |
+
"loss": 0.2916,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2392857142857143,
|
| 475 |
+
"grad_norm": 0.8817863744148133,
|
| 476 |
+
"learning_rate": 1.782347490333123e-05,
|
| 477 |
+
"loss": 0.2599,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.24285714285714285,
|
| 482 |
+
"grad_norm": 0.6848196359492216,
|
| 483 |
+
"learning_rate": 1.775074727505667e-05,
|
| 484 |
+
"loss": 0.2374,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.24642857142857144,
|
| 489 |
+
"grad_norm": 0.8834440854449724,
|
| 490 |
+
"learning_rate": 1.7676978049408262e-05,
|
| 491 |
+
"loss": 0.2134,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.25,
|
| 496 |
+
"grad_norm": 0.8655380809431652,
|
| 497 |
+
"learning_rate": 1.7602177139989046e-05,
|
| 498 |
+
"loss": 0.2456,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.25357142857142856,
|
| 503 |
+
"grad_norm": 0.9506232189055851,
|
| 504 |
+
"learning_rate": 1.7526354599046637e-05,
|
| 505 |
+
"loss": 0.2661,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2571428571428571,
|
| 510 |
+
"grad_norm": 0.7978880305255668,
|
| 511 |
+
"learning_rate": 1.7449520616122344e-05,
|
| 512 |
+
"loss": 0.2355,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.26071428571428573,
|
| 517 |
+
"grad_norm": 0.8250196602876038,
|
| 518 |
+
"learning_rate": 1.7371685516681825e-05,
|
| 519 |
+
"loss": 0.207,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2642857142857143,
|
| 524 |
+
"grad_norm": 0.8117468052334055,
|
| 525 |
+
"learning_rate": 1.7292859760727493e-05,
|
| 526 |
+
"loss": 0.2303,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.26785714285714285,
|
| 531 |
+
"grad_norm": 0.7195318618659722,
|
| 532 |
+
"learning_rate": 1.721305394139282e-05,
|
| 533 |
+
"loss": 0.1655,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2714285714285714,
|
| 538 |
+
"grad_norm": 0.6949862427137922,
|
| 539 |
+
"learning_rate": 1.7132278783518756e-05,
|
| 540 |
+
"loss": 0.2083,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.275,
|
| 545 |
+
"grad_norm": 0.8514564789796272,
|
| 546 |
+
"learning_rate": 1.7050545142212483e-05,
|
| 547 |
+
"loss": 0.2242,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2785714285714286,
|
| 552 |
+
"grad_norm": 0.7516508666701327,
|
| 553 |
+
"learning_rate": 1.696786400138859e-05,
|
| 554 |
+
"loss": 0.2437,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.28214285714285714,
|
| 559 |
+
"grad_norm": 0.9714012683887667,
|
| 560 |
+
"learning_rate": 1.6884246472293018e-05,
|
| 561 |
+
"loss": 0.2375,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2857142857142857,
|
| 566 |
+
"grad_norm": 0.7572091520879666,
|
| 567 |
+
"learning_rate": 1.679970379200983e-05,
|
| 568 |
+
"loss": 0.242,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2892857142857143,
|
| 573 |
+
"grad_norm": 0.8297099751697198,
|
| 574 |
+
"learning_rate": 1.6714247321951106e-05,
|
| 575 |
+
"loss": 0.263,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.29285714285714287,
|
| 580 |
+
"grad_norm": 0.9090423008791377,
|
| 581 |
+
"learning_rate": 1.6627888546330136e-05,
|
| 582 |
+
"loss": 0.2961,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.29642857142857143,
|
| 587 |
+
"grad_norm": 0.8510800428736458,
|
| 588 |
+
"learning_rate": 1.654063907061807e-05,
|
| 589 |
+
"loss": 0.2321,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3,
|
| 594 |
+
"grad_norm": 0.7131209745137069,
|
| 595 |
+
"learning_rate": 1.64525106199843e-05,
|
| 596 |
+
"loss": 0.2288,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.30357142857142855,
|
| 601 |
+
"grad_norm": 0.6919526731968912,
|
| 602 |
+
"learning_rate": 1.6363515037720774e-05,
|
| 603 |
+
"loss": 0.2288,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.30714285714285716,
|
| 608 |
+
"grad_norm": 0.7958343122708701,
|
| 609 |
+
"learning_rate": 1.6273664283650393e-05,
|
| 610 |
+
"loss": 0.2323,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3107142857142857,
|
| 615 |
+
"grad_norm": 0.6483638471772285,
|
| 616 |
+
"learning_rate": 1.6182970432519772e-05,
|
| 617 |
+
"loss": 0.1861,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3142857142857143,
|
| 622 |
+
"grad_norm": 0.7300972630561462,
|
| 623 |
+
"learning_rate": 1.609144567237658e-05,
|
| 624 |
+
"loss": 0.2213,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.31785714285714284,
|
| 629 |
+
"grad_norm": 0.6899792899268994,
|
| 630 |
+
"learning_rate": 1.5999102302931585e-05,
|
| 631 |
+
"loss": 0.2224,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.32142857142857145,
|
| 636 |
+
"grad_norm": 0.767512157488968,
|
| 637 |
+
"learning_rate": 1.5905952733905777e-05,
|
| 638 |
+
"loss": 0.2634,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.325,
|
| 643 |
+
"grad_norm": 0.861421412557911,
|
| 644 |
+
"learning_rate": 1.5812009483362643e-05,
|
| 645 |
+
"loss": 0.2516,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.32857142857142857,
|
| 650 |
+
"grad_norm": 1.2043142385923853,
|
| 651 |
+
"learning_rate": 1.5717285176025913e-05,
|
| 652 |
+
"loss": 0.2389,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.33214285714285713,
|
| 657 |
+
"grad_norm": 1.0812423023626763,
|
| 658 |
+
"learning_rate": 1.5621792541582968e-05,
|
| 659 |
+
"loss": 0.2514,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3357142857142857,
|
| 664 |
+
"grad_norm": 0.8297182880168162,
|
| 665 |
+
"learning_rate": 1.552554441297413e-05,
|
| 666 |
+
"loss": 0.245,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3392857142857143,
|
| 671 |
+
"grad_norm": 0.7460278115544305,
|
| 672 |
+
"learning_rate": 1.5428553724668103e-05,
|
| 673 |
+
"loss": 0.2438,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.34285714285714286,
|
| 678 |
+
"grad_norm": 0.7162063436193922,
|
| 679 |
+
"learning_rate": 1.533083351092372e-05,
|
| 680 |
+
"loss": 0.1747,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3464285714285714,
|
| 685 |
+
"grad_norm": 0.6456220159305923,
|
| 686 |
+
"learning_rate": 1.5232396904038352e-05,
|
| 687 |
+
"loss": 0.1534,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.35,
|
| 692 |
+
"grad_norm": 0.7094654735506336,
|
| 693 |
+
"learning_rate": 1.5133257132583074e-05,
|
| 694 |
+
"loss": 0.2545,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3535714285714286,
|
| 699 |
+
"grad_norm": 0.8247734466190151,
|
| 700 |
+
"learning_rate": 1.503342751962493e-05,
|
| 701 |
+
"loss": 0.2848,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.35714285714285715,
|
| 706 |
+
"grad_norm": 0.7728186991716375,
|
| 707 |
+
"learning_rate": 1.4932921480936491e-05,
|
| 708 |
+
"loss": 0.2566,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3607142857142857,
|
| 713 |
+
"grad_norm": 0.8642822181295595,
|
| 714 |
+
"learning_rate": 1.4831752523192949e-05,
|
| 715 |
+
"loss": 0.2583,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.36428571428571427,
|
| 720 |
+
"grad_norm": 0.8060855146089662,
|
| 721 |
+
"learning_rate": 1.4729934242157005e-05,
|
| 722 |
+
"loss": 0.2754,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3678571428571429,
|
| 727 |
+
"grad_norm": 0.8190230120369192,
|
| 728 |
+
"learning_rate": 1.4627480320851775e-05,
|
| 729 |
+
"loss": 0.2865,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.37142857142857144,
|
| 734 |
+
"grad_norm": 0.6600130809466522,
|
| 735 |
+
"learning_rate": 1.4524404527721977e-05,
|
| 736 |
+
"loss": 0.195,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.375,
|
| 741 |
+
"grad_norm": 0.6884634046078892,
|
| 742 |
+
"learning_rate": 1.4420720714783635e-05,
|
| 743 |
+
"loss": 0.212,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.37857142857142856,
|
| 748 |
+
"grad_norm": 0.7936195003334405,
|
| 749 |
+
"learning_rate": 1.4316442815762543e-05,
|
| 750 |
+
"loss": 0.2658,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3821428571428571,
|
| 755 |
+
"grad_norm": 0.7025320097938863,
|
| 756 |
+
"learning_rate": 1.4211584844221771e-05,
|
| 757 |
+
"loss": 0.2087,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.38571428571428573,
|
| 762 |
+
"grad_norm": 0.6172413774186932,
|
| 763 |
+
"learning_rate": 1.4106160891678422e-05,
|
| 764 |
+
"loss": 0.1727,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3892857142857143,
|
| 769 |
+
"grad_norm": 0.5907547063659139,
|
| 770 |
+
"learning_rate": 1.4000185125709919e-05,
|
| 771 |
+
"loss": 0.1801,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.39285714285714285,
|
| 776 |
+
"grad_norm": 0.7794646496722246,
|
| 777 |
+
"learning_rate": 1.3893671788050073e-05,
|
| 778 |
+
"loss": 0.2018,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3964285714285714,
|
| 783 |
+
"grad_norm": 0.5110684834921772,
|
| 784 |
+
"learning_rate": 1.3786635192675184e-05,
|
| 785 |
+
"loss": 0.1593,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4,
|
| 790 |
+
"grad_norm": 0.9463597697315387,
|
| 791 |
+
"learning_rate": 1.3679089723880427e-05,
|
| 792 |
+
"loss": 0.2699,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.4035714285714286,
|
| 797 |
+
"grad_norm": 0.6376009700059774,
|
| 798 |
+
"learning_rate": 1.35710498343468e-05,
|
| 799 |
+
"loss": 0.1562,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.40714285714285714,
|
| 804 |
+
"grad_norm": 0.7904776444292465,
|
| 805 |
+
"learning_rate": 1.3462530043198874e-05,
|
| 806 |
+
"loss": 0.2702,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.4107142857142857,
|
| 811 |
+
"grad_norm": 0.5896293345349027,
|
| 812 |
+
"learning_rate": 1.3353544934053618e-05,
|
| 813 |
+
"loss": 0.2016,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4142857142857143,
|
| 818 |
+
"grad_norm": 0.6475939911269051,
|
| 819 |
+
"learning_rate": 1.324410915306055e-05,
|
| 820 |
+
"loss": 0.2196,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.41785714285714287,
|
| 825 |
+
"grad_norm": 0.5957948341177419,
|
| 826 |
+
"learning_rate": 1.3134237406933493e-05,
|
| 827 |
+
"loss": 0.1883,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.42142857142857143,
|
| 832 |
+
"grad_norm": 0.7141704868499749,
|
| 833 |
+
"learning_rate": 1.3023944460974183e-05,
|
| 834 |
+
"loss": 0.1765,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.425,
|
| 839 |
+
"grad_norm": 0.6190173244418413,
|
| 840 |
+
"learning_rate": 1.2913245137088024e-05,
|
| 841 |
+
"loss": 0.2338,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.42857142857142855,
|
| 846 |
+
"grad_norm": 0.5152892854036694,
|
| 847 |
+
"learning_rate": 1.2802154311792196e-05,
|
| 848 |
+
"loss": 0.1407,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.43214285714285716,
|
| 853 |
+
"grad_norm": 0.6970455298668206,
|
| 854 |
+
"learning_rate": 1.2690686914216475e-05,
|
| 855 |
+
"loss": 0.2245,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.4357142857142857,
|
| 860 |
+
"grad_norm": 0.5424563727889877,
|
| 861 |
+
"learning_rate": 1.2578857924096935e-05,
|
| 862 |
+
"loss": 0.1642,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4392857142857143,
|
| 867 |
+
"grad_norm": 0.5455879645989501,
|
| 868 |
+
"learning_rate": 1.2466682369762883e-05,
|
| 869 |
+
"loss": 0.1745,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.44285714285714284,
|
| 874 |
+
"grad_norm": 0.7159501319130775,
|
| 875 |
+
"learning_rate": 1.2354175326117252e-05,
|
| 876 |
+
"loss": 0.2147,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.44642857142857145,
|
| 881 |
+
"grad_norm": 0.9747645380400841,
|
| 882 |
+
"learning_rate": 1.2241351912610726e-05,
|
| 883 |
+
"loss": 0.2842,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.45,
|
| 888 |
+
"grad_norm": 0.663151474979954,
|
| 889 |
+
"learning_rate": 1.212822729120989e-05,
|
| 890 |
+
"loss": 0.1917,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.45357142857142857,
|
| 895 |
+
"grad_norm": 0.539525979449454,
|
| 896 |
+
"learning_rate": 1.2014816664359671e-05,
|
| 897 |
+
"loss": 0.1929,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.45714285714285713,
|
| 902 |
+
"grad_norm": 0.7511233269985765,
|
| 903 |
+
"learning_rate": 1.190113527294032e-05,
|
| 904 |
+
"loss": 0.2415,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4607142857142857,
|
| 909 |
+
"grad_norm": 0.6962189495102956,
|
| 910 |
+
"learning_rate": 1.178719839421925e-05,
|
| 911 |
+
"loss": 0.1904,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.4642857142857143,
|
| 916 |
+
"grad_norm": 0.5794963846183635,
|
| 917 |
+
"learning_rate": 1.1673021339797967e-05,
|
| 918 |
+
"loss": 0.2185,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.46785714285714286,
|
| 923 |
+
"grad_norm": 0.5834120344196029,
|
| 924 |
+
"learning_rate": 1.15586194535544e-05,
|
| 925 |
+
"loss": 0.1648,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4714285714285714,
|
| 930 |
+
"grad_norm": 0.42363171997494503,
|
| 931 |
+
"learning_rate": 1.1444008109580884e-05,
|
| 932 |
+
"loss": 0.1274,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.475,
|
| 937 |
+
"grad_norm": 0.7339617718296433,
|
| 938 |
+
"learning_rate": 1.1329202710118088e-05,
|
| 939 |
+
"loss": 0.2408,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4785714285714286,
|
| 944 |
+
"grad_norm": 0.7313570855021271,
|
| 945 |
+
"learning_rate": 1.1214218683485159e-05,
|
| 946 |
+
"loss": 0.2571,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.48214285714285715,
|
| 951 |
+
"grad_norm": 0.6413922491909684,
|
| 952 |
+
"learning_rate": 1.1099071482006361e-05,
|
| 953 |
+
"loss": 0.2049,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4857142857142857,
|
| 958 |
+
"grad_norm": 0.6927763783969573,
|
| 959 |
+
"learning_rate": 1.0983776579934483e-05,
|
| 960 |
+
"loss": 0.2319,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.48928571428571427,
|
| 965 |
+
"grad_norm": 0.8400234609027128,
|
| 966 |
+
"learning_rate": 1.0868349471371316e-05,
|
| 967 |
+
"loss": 0.2984,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4928571428571429,
|
| 972 |
+
"grad_norm": 0.59355253258987,
|
| 973 |
+
"learning_rate": 1.0752805668185442e-05,
|
| 974 |
+
"loss": 0.1682,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.49642857142857144,
|
| 979 |
+
"grad_norm": 0.6702328703095428,
|
| 980 |
+
"learning_rate": 1.0637160697927651e-05,
|
| 981 |
+
"loss": 0.2486,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5,
|
| 986 |
+
"grad_norm": 0.610663540942464,
|
| 987 |
+
"learning_rate": 1.0521430101744238e-05,
|
| 988 |
+
"loss": 0.1944,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5035714285714286,
|
| 993 |
+
"grad_norm": 0.6670778255030475,
|
| 994 |
+
"learning_rate": 1.040562943228849e-05,
|
| 995 |
+
"loss": 0.2303,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5071428571428571,
|
| 1000 |
+
"grad_norm": 0.5298284958723746,
|
| 1001 |
+
"learning_rate": 1.0289774251630602e-05,
|
| 1002 |
+
"loss": 0.1324,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5107142857142857,
|
| 1007 |
+
"grad_norm": 0.7102449237713073,
|
| 1008 |
+
"learning_rate": 1.0173880129166358e-05,
|
| 1009 |
+
"loss": 0.2229,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5142857142857142,
|
| 1014 |
+
"grad_norm": 0.5146460843697134,
|
| 1015 |
+
"learning_rate": 1.0057962639524799e-05,
|
| 1016 |
+
"loss": 0.1508,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5178571428571429,
|
| 1021 |
+
"grad_norm": 0.5757523755087942,
|
| 1022 |
+
"learning_rate": 9.942037360475205e-06,
|
| 1023 |
+
"loss": 0.1728,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5214285714285715,
|
| 1028 |
+
"grad_norm": 0.6500936134711337,
|
| 1029 |
+
"learning_rate": 9.826119870833644e-06,
|
| 1030 |
+
"loss": 0.2437,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.525,
|
| 1035 |
+
"grad_norm": 0.5181133145906637,
|
| 1036 |
+
"learning_rate": 9.710225748369402e-06,
|
| 1037 |
+
"loss": 0.1865,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5285714285714286,
|
| 1042 |
+
"grad_norm": 0.495454322336506,
|
| 1043 |
+
"learning_rate": 9.594370567711512e-06,
|
| 1044 |
+
"loss": 0.1617,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5321428571428571,
|
| 1049 |
+
"grad_norm": 0.6228595031287888,
|
| 1050 |
+
"learning_rate": 9.478569898255765e-06,
|
| 1051 |
+
"loss": 0.2104,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5357142857142857,
|
| 1056 |
+
"grad_norm": 0.44664086678984055,
|
| 1057 |
+
"learning_rate": 9.362839302072354e-06,
|
| 1058 |
+
"loss": 0.1461,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5392857142857143,
|
| 1063 |
+
"grad_norm": 0.6712033221960267,
|
| 1064 |
+
"learning_rate": 9.247194331814561e-06,
|
| 1065 |
+
"loss": 0.2092,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5428571428571428,
|
| 1070 |
+
"grad_norm": 0.5187745738654083,
|
| 1071 |
+
"learning_rate": 9.131650528628688e-06,
|
| 1072 |
+
"loss": 0.1618,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5464285714285714,
|
| 1077 |
+
"grad_norm": 0.5742615521018551,
|
| 1078 |
+
"learning_rate": 9.016223420065519e-06,
|
| 1079 |
+
"loss": 0.1781,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.55,
|
| 1084 |
+
"grad_norm": 0.5996031168106218,
|
| 1085 |
+
"learning_rate": 8.900928517993644e-06,
|
| 1086 |
+
"loss": 0.1895,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5535714285714286,
|
| 1091 |
+
"grad_norm": 0.6328292684226152,
|
| 1092 |
+
"learning_rate": 8.785781316514841e-06,
|
| 1093 |
+
"loss": 0.2149,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5571428571428572,
|
| 1098 |
+
"grad_norm": 0.5387581480854878,
|
| 1099 |
+
"learning_rate": 8.670797289881915e-06,
|
| 1100 |
+
"loss": 0.1775,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5607142857142857,
|
| 1105 |
+
"grad_norm": 0.5572164434404691,
|
| 1106 |
+
"learning_rate": 8.555991890419116e-06,
|
| 1107 |
+
"loss": 0.1933,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5642857142857143,
|
| 1112 |
+
"grad_norm": 0.5649031155667329,
|
| 1113 |
+
"learning_rate": 8.441380546445603e-06,
|
| 1114 |
+
"loss": 0.181,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5678571428571428,
|
| 1119 |
+
"grad_norm": 0.6573973388367828,
|
| 1120 |
+
"learning_rate": 8.326978660202034e-06,
|
| 1121 |
+
"loss": 0.2374,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5714285714285714,
|
| 1126 |
+
"grad_norm": 0.6303978475727942,
|
| 1127 |
+
"learning_rate": 8.212801605780754e-06,
|
| 1128 |
+
"loss": 0.1735,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.575,
|
| 1133 |
+
"grad_norm": 0.4801400244382432,
|
| 1134 |
+
"learning_rate": 8.098864727059685e-06,
|
| 1135 |
+
"loss": 0.1412,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5785714285714286,
|
| 1140 |
+
"grad_norm": 0.6847676384869728,
|
| 1141 |
+
"learning_rate": 7.985183335640332e-06,
|
| 1142 |
+
"loss": 0.2142,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5821428571428572,
|
| 1147 |
+
"grad_norm": 0.42794916312384695,
|
| 1148 |
+
"learning_rate": 7.871772708790114e-06,
|
| 1149 |
+
"loss": 0.1167,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5857142857142857,
|
| 1154 |
+
"grad_norm": 0.7293980682002585,
|
| 1155 |
+
"learning_rate": 7.758648087389277e-06,
|
| 1156 |
+
"loss": 0.2031,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5892857142857143,
|
| 1161 |
+
"grad_norm": 0.6599403699786116,
|
| 1162 |
+
"learning_rate": 7.64582467388275e-06,
|
| 1163 |
+
"loss": 0.2415,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5928571428571429,
|
| 1168 |
+
"grad_norm": 0.48043926105927637,
|
| 1169 |
+
"learning_rate": 7.533317630237117e-06,
|
| 1170 |
+
"loss": 0.1888,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5964285714285714,
|
| 1175 |
+
"grad_norm": 0.6412422605963668,
|
| 1176 |
+
"learning_rate": 7.421142075903067e-06,
|
| 1177 |
+
"loss": 0.2089,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6,
|
| 1182 |
+
"grad_norm": 0.6809868730952757,
|
| 1183 |
+
"learning_rate": 7.3093130857835245e-06,
|
| 1184 |
+
"loss": 0.1805,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6035714285714285,
|
| 1189 |
+
"grad_norm": 0.6418069717776665,
|
| 1190 |
+
"learning_rate": 7.197845688207805e-06,
|
| 1191 |
+
"loss": 0.2438,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6071428571428571,
|
| 1196 |
+
"grad_norm": 0.5376357495185539,
|
| 1197 |
+
"learning_rate": 7.086754862911982e-06,
|
| 1198 |
+
"loss": 0.2239,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6107142857142858,
|
| 1203 |
+
"grad_norm": 0.682482573713786,
|
| 1204 |
+
"learning_rate": 6.976055539025819e-06,
|
| 1205 |
+
"loss": 0.2778,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6142857142857143,
|
| 1210 |
+
"grad_norm": 0.5510808352313203,
|
| 1211 |
+
"learning_rate": 6.865762593066514e-06,
|
| 1212 |
+
"loss": 0.1768,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6178571428571429,
|
| 1217 |
+
"grad_norm": 1.0898121691999565,
|
| 1218 |
+
"learning_rate": 6.755890846939454e-06,
|
| 1219 |
+
"loss": 0.2274,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6214285714285714,
|
| 1224 |
+
"grad_norm": 0.6833115682394764,
|
| 1225 |
+
"learning_rate": 6.646455065946386e-06,
|
| 1226 |
+
"loss": 0.209,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.625,
|
| 1231 |
+
"grad_norm": 0.5834718836130244,
|
| 1232 |
+
"learning_rate": 6.537469956801128e-06,
|
| 1233 |
+
"loss": 0.195,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6285714285714286,
|
| 1238 |
+
"grad_norm": 0.4059918209883239,
|
| 1239 |
+
"learning_rate": 6.428950165653204e-06,
|
| 1240 |
+
"loss": 0.1253,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6321428571428571,
|
| 1245 |
+
"grad_norm": 0.6159299376512053,
|
| 1246 |
+
"learning_rate": 6.320910276119576e-06,
|
| 1247 |
+
"loss": 0.2089,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6357142857142857,
|
| 1252 |
+
"grad_norm": 0.5708201106967846,
|
| 1253 |
+
"learning_rate": 6.213364807324817e-06,
|
| 1254 |
+
"loss": 0.1836,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6392857142857142,
|
| 1259 |
+
"grad_norm": 0.46753041275070745,
|
| 1260 |
+
"learning_rate": 6.106328211949928e-06,
|
| 1261 |
+
"loss": 0.151,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6428571428571429,
|
| 1266 |
+
"grad_norm": 0.46419983181684327,
|
| 1267 |
+
"learning_rate": 5.999814874290084e-06,
|
| 1268 |
+
"loss": 0.1424,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6464285714285715,
|
| 1273 |
+
"grad_norm": 0.6470305272135881,
|
| 1274 |
+
"learning_rate": 5.893839108321584e-06,
|
| 1275 |
+
"loss": 0.2592,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.65,
|
| 1280 |
+
"grad_norm": 0.582995072151267,
|
| 1281 |
+
"learning_rate": 5.7884151557782305e-06,
|
| 1282 |
+
"loss": 0.1394,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6535714285714286,
|
| 1287 |
+
"grad_norm": 0.6872272149230029,
|
| 1288 |
+
"learning_rate": 5.68355718423746e-06,
|
| 1289 |
+
"loss": 0.2273,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6571428571428571,
|
| 1294 |
+
"grad_norm": 0.7010753300233663,
|
| 1295 |
+
"learning_rate": 5.579279285216369e-06,
|
| 1296 |
+
"loss": 0.2578,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6607142857142857,
|
| 1301 |
+
"grad_norm": 0.6816215511289796,
|
| 1302 |
+
"learning_rate": 5.4755954722780236e-06,
|
| 1303 |
+
"loss": 0.2334,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6642857142857143,
|
| 1308 |
+
"grad_norm": 0.9305267527410732,
|
| 1309 |
+
"learning_rate": 5.372519679148227e-06,
|
| 1310 |
+
"loss": 0.3146,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6678571428571428,
|
| 1315 |
+
"grad_norm": 0.5828749443747008,
|
| 1316 |
+
"learning_rate": 5.270065757843e-06,
|
| 1317 |
+
"loss": 0.2424,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6714285714285714,
|
| 1322 |
+
"grad_norm": 0.7646561775047477,
|
| 1323 |
+
"learning_rate": 5.168247476807054e-06,
|
| 1324 |
+
"loss": 0.2737,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.675,
|
| 1329 |
+
"grad_norm": 0.5444525062063674,
|
| 1330 |
+
"learning_rate": 5.067078519063514e-06,
|
| 1331 |
+
"loss": 0.1996,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6785714285714286,
|
| 1336 |
+
"grad_norm": 0.5755758677832743,
|
| 1337 |
+
"learning_rate": 4.966572480375076e-06,
|
| 1338 |
+
"loss": 0.1933,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6821428571428572,
|
| 1343 |
+
"grad_norm": 0.6848265137867346,
|
| 1344 |
+
"learning_rate": 4.86674286741693e-06,
|
| 1345 |
+
"loss": 0.2527,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6857142857142857,
|
| 1350 |
+
"grad_norm": 0.8336972713921105,
|
| 1351 |
+
"learning_rate": 4.767603095961652e-06,
|
| 1352 |
+
"loss": 0.2904,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6892857142857143,
|
| 1357 |
+
"grad_norm": 0.7067542980447211,
|
| 1358 |
+
"learning_rate": 4.669166489076283e-06,
|
| 1359 |
+
"loss": 0.2754,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6928571428571428,
|
| 1364 |
+
"grad_norm": 0.5057729355169331,
|
| 1365 |
+
"learning_rate": 4.571446275331903e-06,
|
| 1366 |
+
"loss": 0.1903,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6964285714285714,
|
| 1371 |
+
"grad_norm": 0.6302783329150187,
|
| 1372 |
+
"learning_rate": 4.47445558702587e-06,
|
| 1373 |
+
"loss": 0.2433,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7,
|
| 1378 |
+
"grad_norm": 0.7066408928533802,
|
| 1379 |
+
"learning_rate": 4.378207458417035e-06,
|
| 1380 |
+
"loss": 0.2295,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7035714285714286,
|
| 1385 |
+
"grad_norm": 0.5840045137089293,
|
| 1386 |
+
"learning_rate": 4.282714823974088e-06,
|
| 1387 |
+
"loss": 0.1797,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.7071428571428572,
|
| 1392 |
+
"grad_norm": 0.5993859301367765,
|
| 1393 |
+
"learning_rate": 4.187990516637361e-06,
|
| 1394 |
+
"loss": 0.1884,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7107142857142857,
|
| 1399 |
+
"grad_norm": 0.5400595981826822,
|
| 1400 |
+
"learning_rate": 4.094047266094225e-06,
|
| 1401 |
+
"loss": 0.177,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7142857142857143,
|
| 1406 |
+
"grad_norm": 0.3473237225491981,
|
| 1407 |
+
"learning_rate": 4.000897697068418e-06,
|
| 1408 |
+
"loss": 0.1096,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7178571428571429,
|
| 1413 |
+
"grad_norm": 0.5910893499694284,
|
| 1414 |
+
"learning_rate": 3.908554327623425e-06,
|
| 1415 |
+
"loss": 0.228,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7214285714285714,
|
| 1420 |
+
"grad_norm": 0.60398021763633,
|
| 1421 |
+
"learning_rate": 3.817029567480228e-06,
|
| 1422 |
+
"loss": 0.2608,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.725,
|
| 1427 |
+
"grad_norm": 0.6200500316907418,
|
| 1428 |
+
"learning_rate": 3.7263357163496118e-06,
|
| 1429 |
+
"loss": 0.2154,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7285714285714285,
|
| 1434 |
+
"grad_norm": 0.5380162414775479,
|
| 1435 |
+
"learning_rate": 3.6364849622792262e-06,
|
| 1436 |
+
"loss": 0.1595,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.7321428571428571,
|
| 1441 |
+
"grad_norm": 0.7246641798143522,
|
| 1442 |
+
"learning_rate": 3.5474893800157005e-06,
|
| 1443 |
+
"loss": 0.2228,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7357142857142858,
|
| 1448 |
+
"grad_norm": 0.45508651388658605,
|
| 1449 |
+
"learning_rate": 3.459360929381931e-06,
|
| 1450 |
+
"loss": 0.1591,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.7392857142857143,
|
| 1455 |
+
"grad_norm": 0.6429141963072342,
|
| 1456 |
+
"learning_rate": 3.372111453669864e-06,
|
| 1457 |
+
"loss": 0.2403,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7428571428571429,
|
| 1462 |
+
"grad_norm": 0.5855029330521654,
|
| 1463 |
+
"learning_rate": 3.2857526780488925e-06,
|
| 1464 |
+
"loss": 0.1934,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7464285714285714,
|
| 1469 |
+
"grad_norm": 0.5125660874685503,
|
| 1470 |
+
"learning_rate": 3.2002962079901743e-06,
|
| 1471 |
+
"loss": 0.1553,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.75,
|
| 1476 |
+
"grad_norm": 0.6976778978872136,
|
| 1477 |
+
"learning_rate": 3.115753527706986e-06,
|
| 1478 |
+
"loss": 0.2384,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7535714285714286,
|
| 1483 |
+
"grad_norm": 0.47434866039192386,
|
| 1484 |
+
"learning_rate": 3.0321359986114096e-06,
|
| 1485 |
+
"loss": 0.1791,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7571428571428571,
|
| 1490 |
+
"grad_norm": 0.514888686551834,
|
| 1491 |
+
"learning_rate": 2.9494548577875195e-06,
|
| 1492 |
+
"loss": 0.1497,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7607142857142857,
|
| 1497 |
+
"grad_norm": 0.5887635651830888,
|
| 1498 |
+
"learning_rate": 2.8677212164812464e-06,
|
| 1499 |
+
"loss": 0.184,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7642857142857142,
|
| 1504 |
+
"grad_norm": 0.5947541481450019,
|
| 1505 |
+
"learning_rate": 2.786946058607187e-06,
|
| 1506 |
+
"loss": 0.1876,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7678571428571429,
|
| 1511 |
+
"grad_norm": 0.7110105563805584,
|
| 1512 |
+
"learning_rate": 2.70714023927251e-06,
|
| 1513 |
+
"loss": 0.2471,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7714285714285715,
|
| 1518 |
+
"grad_norm": 0.6416256915388147,
|
| 1519 |
+
"learning_rate": 2.628314483318178e-06,
|
| 1520 |
+
"loss": 0.2392,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.775,
|
| 1525 |
+
"grad_norm": 0.5903622152845527,
|
| 1526 |
+
"learning_rate": 2.5504793838776585e-06,
|
| 1527 |
+
"loss": 0.1842,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7785714285714286,
|
| 1532 |
+
"grad_norm": 0.6929636779116681,
|
| 1533 |
+
"learning_rate": 2.473645400953366e-06,
|
| 1534 |
+
"loss": 0.2768,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7821428571428571,
|
| 1539 |
+
"grad_norm": 0.5903013240273918,
|
| 1540 |
+
"learning_rate": 2.3978228600109564e-06,
|
| 1541 |
+
"loss": 0.2091,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7857142857142857,
|
| 1546 |
+
"grad_norm": 0.624843633893901,
|
| 1547 |
+
"learning_rate": 2.323021950591743e-06,
|
| 1548 |
+
"loss": 0.2183,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7892857142857143,
|
| 1553 |
+
"grad_norm": 0.626404755888219,
|
| 1554 |
+
"learning_rate": 2.249252724943336e-06,
|
| 1555 |
+
"loss": 0.236,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7928571428571428,
|
| 1560 |
+
"grad_norm": 0.6531550403690458,
|
| 1561 |
+
"learning_rate": 2.176525096668769e-06,
|
| 1562 |
+
"loss": 0.249,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7964285714285714,
|
| 1567 |
+
"grad_norm": 0.6861326233839365,
|
| 1568 |
+
"learning_rate": 2.1048488393942455e-06,
|
| 1569 |
+
"loss": 0.2167,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.8,
|
| 1574 |
+
"grad_norm": 0.5361387918675531,
|
| 1575 |
+
"learning_rate": 2.0342335854556738e-06,
|
| 1576 |
+
"loss": 0.2032,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.8035714285714286,
|
| 1581 |
+
"grad_norm": 0.5567786640700992,
|
| 1582 |
+
"learning_rate": 1.964688824604234e-06,
|
| 1583 |
+
"loss": 0.2286,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8071428571428572,
|
| 1588 |
+
"grad_norm": 0.44674720095893333,
|
| 1589 |
+
"learning_rate": 1.896223902731058e-06,
|
| 1590 |
+
"loss": 0.1147,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8107142857142857,
|
| 1595 |
+
"grad_norm": 0.41283033472542874,
|
| 1596 |
+
"learning_rate": 1.8288480206112879e-06,
|
| 1597 |
+
"loss": 0.1339,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8142857142857143,
|
| 1602 |
+
"grad_norm": 0.7296847675489324,
|
| 1603 |
+
"learning_rate": 1.7625702326675952e-06,
|
| 1604 |
+
"loss": 0.2027,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.8178571428571428,
|
| 1609 |
+
"grad_norm": 0.5271072364545172,
|
| 1610 |
+
"learning_rate": 1.6973994457534026e-06,
|
| 1611 |
+
"loss": 0.1712,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8214285714285714,
|
| 1616 |
+
"grad_norm": 0.672253098189724,
|
| 1617 |
+
"learning_rate": 1.6333444179559078e-06,
|
| 1618 |
+
"loss": 0.2548,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.825,
|
| 1623 |
+
"grad_norm": 0.6715310371949257,
|
| 1624 |
+
"learning_rate": 1.5704137574191202e-06,
|
| 1625 |
+
"loss": 0.2507,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8285714285714286,
|
| 1630 |
+
"grad_norm": 0.6242146450408651,
|
| 1631 |
+
"learning_rate": 1.5086159211870445e-06,
|
| 1632 |
+
"loss": 0.2378,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.8321428571428572,
|
| 1637 |
+
"grad_norm": 0.8815010954927127,
|
| 1638 |
+
"learning_rate": 1.447959214067155e-06,
|
| 1639 |
+
"loss": 0.2415,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.8357142857142857,
|
| 1644 |
+
"grad_norm": 0.563249001616026,
|
| 1645 |
+
"learning_rate": 1.3884517875143544e-06,
|
| 1646 |
+
"loss": 0.188,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.8392857142857143,
|
| 1651 |
+
"grad_norm": 0.6913123639322339,
|
| 1652 |
+
"learning_rate": 1.3301016385355093e-06,
|
| 1653 |
+
"loss": 0.2719,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8428571428571429,
|
| 1658 |
+
"grad_norm": 0.40615590184512884,
|
| 1659 |
+
"learning_rate": 1.2729166086147803e-06,
|
| 1660 |
+
"loss": 0.1136,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8464285714285714,
|
| 1665 |
+
"grad_norm": 0.5857754293443003,
|
| 1666 |
+
"learning_rate": 1.216904382659806e-06,
|
| 1667 |
+
"loss": 0.234,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.85,
|
| 1672 |
+
"grad_norm": 0.823607304030527,
|
| 1673 |
+
"learning_rate": 1.1620724879689793e-06,
|
| 1674 |
+
"loss": 0.2809,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8535714285714285,
|
| 1679 |
+
"grad_norm": 0.6169005264616082,
|
| 1680 |
+
"learning_rate": 1.1084282932198543e-06,
|
| 1681 |
+
"loss": 0.1763,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8571428571428571,
|
| 1686 |
+
"grad_norm": 0.6682010561482211,
|
| 1687 |
+
"learning_rate": 1.0559790074789134e-06,
|
| 1688 |
+
"loss": 0.2398,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8607142857142858,
|
| 1693 |
+
"grad_norm": 0.5418866168713049,
|
| 1694 |
+
"learning_rate": 1.00473167923275e-06,
|
| 1695 |
+
"loss": 0.1964,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8642857142857143,
|
| 1700 |
+
"grad_norm": 0.7106090644853337,
|
| 1701 |
+
"learning_rate": 9.546931954408622e-07,
|
| 1702 |
+
"loss": 0.2337,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8678571428571429,
|
| 1707 |
+
"grad_norm": 0.7417349503796012,
|
| 1708 |
+
"learning_rate": 9.058702806101172e-07,
|
| 1709 |
+
"loss": 0.2632,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8714285714285714,
|
| 1714 |
+
"grad_norm": 0.5175232713000997,
|
| 1715 |
+
"learning_rate": 8.582694958910809e-07,
|
| 1716 |
+
"loss": 0.193,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.875,
|
| 1721 |
+
"grad_norm": 1.026575312488047,
|
| 1722 |
+
"learning_rate": 8.118972381962853e-07,
|
| 1723 |
+
"loss": 0.3553,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8785714285714286,
|
| 1728 |
+
"grad_norm": 0.6308180674164335,
|
| 1729 |
+
"learning_rate": 7.667597393405602e-07,
|
| 1730 |
+
"loss": 0.1723,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8821428571428571,
|
| 1735 |
+
"grad_norm": 0.4590057669806479,
|
| 1736 |
+
"learning_rate": 7.228630652035717e-07,
|
| 1737 |
+
"loss": 0.1531,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8857142857142857,
|
| 1742 |
+
"grad_norm": 0.666809146038292,
|
| 1743 |
+
"learning_rate": 6.802131149146374e-07,
|
| 1744 |
+
"loss": 0.2449,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8892857142857142,
|
| 1749 |
+
"grad_norm": 0.5482805091255768,
|
| 1750 |
+
"learning_rate": 6.388156200599726e-07,
|
| 1751 |
+
"loss": 0.1861,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8928571428571429,
|
| 1756 |
+
"grad_norm": 0.5750571040464779,
|
| 1757 |
+
"learning_rate": 5.986761439124289e-07,
|
| 1758 |
+
"loss": 0.2074,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8964285714285715,
|
| 1763 |
+
"grad_norm": 0.8750045040878164,
|
| 1764 |
+
"learning_rate": 5.598000806838766e-07,
|
| 1765 |
+
"loss": 0.3201,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.9,
|
| 1770 |
+
"grad_norm": 0.48638037681505203,
|
| 1771 |
+
"learning_rate": 5.221926548002876e-07,
|
| 1772 |
+
"loss": 0.1644,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.9035714285714286,
|
| 1777 |
+
"grad_norm": 0.49478576603633184,
|
| 1778 |
+
"learning_rate": 4.858589201996433e-07,
|
| 1779 |
+
"loss": 0.1918,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9071428571428571,
|
| 1784 |
+
"grad_norm": 0.598927975154658,
|
| 1785 |
+
"learning_rate": 4.5080375965275256e-07,
|
| 1786 |
+
"loss": 0.1711,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9107142857142857,
|
| 1791 |
+
"grad_norm": 0.6197004195423823,
|
| 1792 |
+
"learning_rate": 4.1703188410707087e-07,
|
| 1793 |
+
"loss": 0.1928,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.9142857142857143,
|
| 1798 |
+
"grad_norm": 0.6313460016595605,
|
| 1799 |
+
"learning_rate": 3.845478320536178e-07,
|
| 1800 |
+
"loss": 0.2549,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.9178571428571428,
|
| 1805 |
+
"grad_norm": 0.8956056686691475,
|
| 1806 |
+
"learning_rate": 3.5335596891705406e-07,
|
| 1807 |
+
"loss": 0.2841,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.9214285714285714,
|
| 1812 |
+
"grad_norm": 0.4910383596976099,
|
| 1813 |
+
"learning_rate": 3.2346048646903494e-07,
|
| 1814 |
+
"loss": 0.1827,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.925,
|
| 1819 |
+
"grad_norm": 0.6510508585080252,
|
| 1820 |
+
"learning_rate": 2.9486540226488556e-07,
|
| 1821 |
+
"loss": 0.2415,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9285714285714286,
|
| 1826 |
+
"grad_norm": 0.7480168465217824,
|
| 1827 |
+
"learning_rate": 2.6757455910370486e-07,
|
| 1828 |
+
"loss": 0.2374,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.9321428571428572,
|
| 1833 |
+
"grad_norm": 0.47073582050435153,
|
| 1834 |
+
"learning_rate": 2.41591624511931e-07,
|
| 1835 |
+
"loss": 0.1982,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.9357142857142857,
|
| 1840 |
+
"grad_norm": 0.5954530106280657,
|
| 1841 |
+
"learning_rate": 2.169200902504842e-07,
|
| 1842 |
+
"loss": 0.2321,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.9392857142857143,
|
| 1847 |
+
"grad_norm": 0.5083864479298373,
|
| 1848 |
+
"learning_rate": 1.9356327184551716e-07,
|
| 1849 |
+
"loss": 0.2056,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.9428571428571428,
|
| 1854 |
+
"grad_norm": 0.6176017928735918,
|
| 1855 |
+
"learning_rate": 1.7152430814285303e-07,
|
| 1856 |
+
"loss": 0.2248,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9464285714285714,
|
| 1861 |
+
"grad_norm": 0.8794551027561036,
|
| 1862 |
+
"learning_rate": 1.5080616088616884e-07,
|
| 1863 |
+
"loss": 0.276,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.95,
|
| 1868 |
+
"grad_norm": 0.5963379475552784,
|
| 1869 |
+
"learning_rate": 1.3141161431896809e-07,
|
| 1870 |
+
"loss": 0.2682,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9535714285714286,
|
| 1875 |
+
"grad_norm": 0.6298840458046066,
|
| 1876 |
+
"learning_rate": 1.1334327481042573e-07,
|
| 1877 |
+
"loss": 0.1362,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.9571428571428572,
|
| 1882 |
+
"grad_norm": 0.5539259365985325,
|
| 1883 |
+
"learning_rate": 9.660357050512158e-08,
|
| 1884 |
+
"loss": 0.2111,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9607142857142857,
|
| 1889 |
+
"grad_norm": 0.7343899826298071,
|
| 1890 |
+
"learning_rate": 8.119475099673035e-08,
|
| 1891 |
+
"loss": 0.2867,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9642857142857143,
|
| 1896 |
+
"grad_norm": 0.5275488999179762,
|
| 1897 |
+
"learning_rate": 6.711888702570556e-08,
|
| 1898 |
+
"loss": 0.1578,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9678571428571429,
|
| 1903 |
+
"grad_norm": 0.5471929852386093,
|
| 1904 |
+
"learning_rate": 5.437787020100116e-08,
|
| 1905 |
+
"loss": 0.1831,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9714285714285714,
|
| 1910 |
+
"grad_norm": 0.7137357035927524,
|
| 1911 |
+
"learning_rate": 4.297341274586475e-08,
|
| 1912 |
+
"loss": 0.2711,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.975,
|
| 1917 |
+
"grad_norm": 0.6861484126427937,
|
| 1918 |
+
"learning_rate": 3.290704726773619e-08,
|
| 1919 |
+
"loss": 0.2091,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9785714285714285,
|
| 1924 |
+
"grad_norm": 0.7660872222264061,
|
| 1925 |
+
"learning_rate": 2.4180126552284523e-08,
|
| 1926 |
+
"loss": 0.2517,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.9821428571428571,
|
| 1931 |
+
"grad_norm": 0.6124965928901931,
|
| 1932 |
+
"learning_rate": 1.6793823381614506e-08,
|
| 1933 |
+
"loss": 0.2005,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.9857142857142858,
|
| 1938 |
+
"grad_norm": 0.6082415014874585,
|
| 1939 |
+
"learning_rate": 1.0749130376659366e-08,
|
| 1940 |
+
"loss": 0.224,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9892857142857143,
|
| 1945 |
+
"grad_norm": 0.6973577336250109,
|
| 1946 |
+
"learning_rate": 6.046859863781951e-09,
|
| 1947 |
+
"loss": 0.2659,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9928571428571429,
|
| 1952 |
+
"grad_norm": 0.6364613016078383,
|
| 1953 |
+
"learning_rate": 2.6876437656153663e-09,
|
| 1954 |
+
"loss": 0.2133,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.9964285714285714,
|
| 1959 |
+
"grad_norm": 0.5410559341159564,
|
| 1960 |
+
"learning_rate": 6.719335161364804e-10,
|
| 1961 |
+
"loss": 0.1768,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 1.0,
|
| 1966 |
+
"grad_norm": 0.684252146873541,
|
| 1967 |
+
"learning_rate": 0.0,
|
| 1968 |
+
"loss": 0.1917,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 1.0,
|
| 1973 |
+
"step": 280,
|
| 1974 |
+
"total_flos": 222415492415488.0,
|
| 1975 |
+
"train_loss": 0.3071911358673658,
|
| 1976 |
+
"train_runtime": 1347.2293,
|
| 1977 |
+
"train_samples_per_second": 26.594,
|
| 1978 |
+
"train_steps_per_second": 0.208
|
| 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": 222415492415488.0,
|
| 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 sha256:10cb07e544774f13dc6a645be9dc41bb1aa44c6da9d379dd232f3d32288eeb4d
|
| 3 |
+
size 7224
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|