Add files using upload-large-folder tool
Browse files- added_tokens.json +37 -0
- chat_template.jinja +6 -0
- config.json +123 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +40 -0
- tokenizer_config.json +322 -0
- trainer_state.json +934 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</box>": 151677,
|
| 3 |
+
"</img>": 151670,
|
| 4 |
+
"</quad>": 151673,
|
| 5 |
+
"</ref>": 151675,
|
| 6 |
+
"</think>": 151668,
|
| 7 |
+
"</tool_call>": 151658,
|
| 8 |
+
"</tool_response>": 151666,
|
| 9 |
+
"<IMG_CONTEXT>": 151671,
|
| 10 |
+
"<box>": 151676,
|
| 11 |
+
"<img>": 151669,
|
| 12 |
+
"<quad>": 151672,
|
| 13 |
+
"<ref>": 151674,
|
| 14 |
+
"<think>": 151667,
|
| 15 |
+
"<tool_call>": 151657,
|
| 16 |
+
"<tool_response>": 151665,
|
| 17 |
+
"<|box_end|>": 151649,
|
| 18 |
+
"<|box_start|>": 151648,
|
| 19 |
+
"<|endoftext|>": 151643,
|
| 20 |
+
"<|file_sep|>": 151664,
|
| 21 |
+
"<|fim_middle|>": 151660,
|
| 22 |
+
"<|fim_pad|>": 151662,
|
| 23 |
+
"<|fim_prefix|>": 151659,
|
| 24 |
+
"<|fim_suffix|>": 151661,
|
| 25 |
+
"<|im_end|>": 151645,
|
| 26 |
+
"<|im_start|>": 151644,
|
| 27 |
+
"<|image_pad|>": 151655,
|
| 28 |
+
"<|object_ref_end|>": 151647,
|
| 29 |
+
"<|object_ref_start|>": 151646,
|
| 30 |
+
"<|quad_end|>": 151651,
|
| 31 |
+
"<|quad_start|>": 151650,
|
| 32 |
+
"<|repo_name|>": 151663,
|
| 33 |
+
"<|video_pad|>": 151656,
|
| 34 |
+
"<|vision_end|>": 151653,
|
| 35 |
+
"<|vision_pad|>": 151654,
|
| 36 |
+
"<|vision_start|>": 151652
|
| 37 |
+
}
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{{'<|im_start|>' + message['role'] + '
|
| 2 |
+
'}}{% if message['content'] is string %}{{ message['content'] }}{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' %}{{ '<image>
|
| 3 |
+
' }}{% elif content['type'] == 'video' %}{{ '<video>
|
| 4 |
+
' }}{% elif content['type'] == 'text' %}{{ content['text'] }}{% endif %}{% endfor %}{% endif %}{{'<|im_end|>
|
| 5 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{'<|im_start|>assistant
|
| 6 |
+
' }}{% endif %}
|
config.json
ADDED
|
@@ -0,0 +1,123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"InternVLChatModel"
|
| 4 |
+
],
|
| 5 |
+
"auto_map": {
|
| 6 |
+
"AutoConfig": "configuration_internvl_chat.InternVLChatConfig",
|
| 7 |
+
"AutoModel": "modeling_internvl_chat.InternVLChatModel",
|
| 8 |
+
"AutoModelForCausalLM": "modeling_internvl_chat.InternVLChatModel"
|
| 9 |
+
},
|
| 10 |
+
"downsample_ratio": 0.5,
|
| 11 |
+
"dtype": "bfloat16",
|
| 12 |
+
"dynamic_image_size": true,
|
| 13 |
+
"eos_token_id": 151645,
|
| 14 |
+
"force_image_size": 336,
|
| 15 |
+
"llm_config": {
|
| 16 |
+
"_name_or_path": "/root/codespace/checkpoints/Qwen3-1.7B",
|
| 17 |
+
"architectures": [
|
| 18 |
+
"Qwen3ForCausalLM"
|
| 19 |
+
],
|
| 20 |
+
"attention_bias": false,
|
| 21 |
+
"attention_dropout": 0.0,
|
| 22 |
+
"bos_token_id": 151643,
|
| 23 |
+
"debug": false,
|
| 24 |
+
"dtype": "bfloat16",
|
| 25 |
+
"eos_token_id": 151645,
|
| 26 |
+
"ep_size": 1,
|
| 27 |
+
"head_dim": 128,
|
| 28 |
+
"hidden_act": "silu",
|
| 29 |
+
"hidden_size": 2048,
|
| 30 |
+
"initializer_range": 0.02,
|
| 31 |
+
"intermediate_size": 6144,
|
| 32 |
+
"layer_types": [
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"full_attention",
|
| 54 |
+
"full_attention",
|
| 55 |
+
"full_attention",
|
| 56 |
+
"full_attention",
|
| 57 |
+
"full_attention",
|
| 58 |
+
"full_attention",
|
| 59 |
+
"full_attention",
|
| 60 |
+
"full_attention"
|
| 61 |
+
],
|
| 62 |
+
"max_position_embeddings": 40960,
|
| 63 |
+
"max_window_layers": 28,
|
| 64 |
+
"micro_forward": false,
|
| 65 |
+
"model_type": "qwen3",
|
| 66 |
+
"num_attention_heads": 16,
|
| 67 |
+
"num_hidden_layers": 28,
|
| 68 |
+
"num_key_value_heads": 8,
|
| 69 |
+
"rms_norm_eps": 1e-06,
|
| 70 |
+
"rope_scaling": null,
|
| 71 |
+
"rope_theta": 1000000,
|
| 72 |
+
"skip_checkpoint": false,
|
| 73 |
+
"sliding_window": null,
|
| 74 |
+
"use_cache": false,
|
| 75 |
+
"use_deepep": false,
|
| 76 |
+
"use_sliding_window": false,
|
| 77 |
+
"vocab_size": 151936
|
| 78 |
+
},
|
| 79 |
+
"max_dynamic_patch": 12,
|
| 80 |
+
"min_dynamic_patch": 1,
|
| 81 |
+
"model_type": "internvl_chat",
|
| 82 |
+
"output_attentions": false,
|
| 83 |
+
"pad2square": false,
|
| 84 |
+
"pad_token_id": 151643,
|
| 85 |
+
"ps_version": "v2",
|
| 86 |
+
"select_layer": -1,
|
| 87 |
+
"template": "internvl2_5",
|
| 88 |
+
"tie_word_embeddings": false,
|
| 89 |
+
"transformers_version": null,
|
| 90 |
+
"use_backbone_lora": 0,
|
| 91 |
+
"use_llm_lora": 0,
|
| 92 |
+
"use_thumbnail": true,
|
| 93 |
+
"vision_config": {
|
| 94 |
+
"architectures": [
|
| 95 |
+
"InternVisionModel"
|
| 96 |
+
],
|
| 97 |
+
"attention_dropout": 0.0,
|
| 98 |
+
"auto_map": {
|
| 99 |
+
"AutoConfig": "configuration_intern_vit.InternVisionConfig",
|
| 100 |
+
"AutoModel": "modeling_intern_vit.InternVisionModel"
|
| 101 |
+
},
|
| 102 |
+
"drop_path_rate": 0.0,
|
| 103 |
+
"dropout": 0.0,
|
| 104 |
+
"dtype": "bfloat16",
|
| 105 |
+
"hidden_act": "gelu",
|
| 106 |
+
"hidden_size": 1024,
|
| 107 |
+
"image_size": 448,
|
| 108 |
+
"initializer_factor": 1.0,
|
| 109 |
+
"initializer_range": 0.02,
|
| 110 |
+
"intermediate_size": 4096,
|
| 111 |
+
"layer_norm_eps": 1e-06,
|
| 112 |
+
"model_type": "intern_vit_6b",
|
| 113 |
+
"norm_type": "layer_norm",
|
| 114 |
+
"num_attention_heads": 16,
|
| 115 |
+
"num_channels": 3,
|
| 116 |
+
"num_hidden_layers": 24,
|
| 117 |
+
"patch_size": 14,
|
| 118 |
+
"qk_normalization": false,
|
| 119 |
+
"qkv_bias": true,
|
| 120 |
+
"use_fa3": false,
|
| 121 |
+
"use_flash_attn": false
|
| 122 |
+
}
|
| 123 |
+
}
|
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:d5278b8b6ee019fc65a64ba3271f29de60417aac19e6b91e8c6d497d153741cb
|
| 3 |
+
size 4092735128
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>",
|
| 16 |
+
"<img>",
|
| 17 |
+
"</img>",
|
| 18 |
+
"<IMG_CONTEXT>",
|
| 19 |
+
"<quad>",
|
| 20 |
+
"</quad>",
|
| 21 |
+
"<ref>",
|
| 22 |
+
"</ref>",
|
| 23 |
+
"<box>",
|
| 24 |
+
"</box>"
|
| 25 |
+
],
|
| 26 |
+
"eos_token": {
|
| 27 |
+
"content": "<|im_end|>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": false,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
},
|
| 33 |
+
"pad_token": {
|
| 34 |
+
"content": "<|endoftext|>",
|
| 35 |
+
"lstrip": false,
|
| 36 |
+
"normalized": false,
|
| 37 |
+
"rstrip": false,
|
| 38 |
+
"single_word": false
|
| 39 |
+
}
|
| 40 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,322 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": false,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"151643": {
|
| 7 |
+
"content": "<|endoftext|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"151644": {
|
| 15 |
+
"content": "<|im_start|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"151645": {
|
| 23 |
+
"content": "<|im_end|>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"151646": {
|
| 31 |
+
"content": "<|object_ref_start|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"151647": {
|
| 39 |
+
"content": "<|object_ref_end|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
},
|
| 46 |
+
"151648": {
|
| 47 |
+
"content": "<|box_start|>",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": false,
|
| 50 |
+
"rstrip": false,
|
| 51 |
+
"single_word": false,
|
| 52 |
+
"special": true
|
| 53 |
+
},
|
| 54 |
+
"151649": {
|
| 55 |
+
"content": "<|box_end|>",
|
| 56 |
+
"lstrip": false,
|
| 57 |
+
"normalized": false,
|
| 58 |
+
"rstrip": false,
|
| 59 |
+
"single_word": false,
|
| 60 |
+
"special": true
|
| 61 |
+
},
|
| 62 |
+
"151650": {
|
| 63 |
+
"content": "<|quad_start|>",
|
| 64 |
+
"lstrip": false,
|
| 65 |
+
"normalized": false,
|
| 66 |
+
"rstrip": false,
|
| 67 |
+
"single_word": false,
|
| 68 |
+
"special": true
|
| 69 |
+
},
|
| 70 |
+
"151651": {
|
| 71 |
+
"content": "<|quad_end|>",
|
| 72 |
+
"lstrip": false,
|
| 73 |
+
"normalized": false,
|
| 74 |
+
"rstrip": false,
|
| 75 |
+
"single_word": false,
|
| 76 |
+
"special": true
|
| 77 |
+
},
|
| 78 |
+
"151652": {
|
| 79 |
+
"content": "<|vision_start|>",
|
| 80 |
+
"lstrip": false,
|
| 81 |
+
"normalized": false,
|
| 82 |
+
"rstrip": false,
|
| 83 |
+
"single_word": false,
|
| 84 |
+
"special": true
|
| 85 |
+
},
|
| 86 |
+
"151653": {
|
| 87 |
+
"content": "<|vision_end|>",
|
| 88 |
+
"lstrip": false,
|
| 89 |
+
"normalized": false,
|
| 90 |
+
"rstrip": false,
|
| 91 |
+
"single_word": false,
|
| 92 |
+
"special": true
|
| 93 |
+
},
|
| 94 |
+
"151654": {
|
| 95 |
+
"content": "<|vision_pad|>",
|
| 96 |
+
"lstrip": false,
|
| 97 |
+
"normalized": false,
|
| 98 |
+
"rstrip": false,
|
| 99 |
+
"single_word": false,
|
| 100 |
+
"special": true
|
| 101 |
+
},
|
| 102 |
+
"151655": {
|
| 103 |
+
"content": "<|image_pad|>",
|
| 104 |
+
"lstrip": false,
|
| 105 |
+
"normalized": false,
|
| 106 |
+
"rstrip": false,
|
| 107 |
+
"single_word": false,
|
| 108 |
+
"special": true
|
| 109 |
+
},
|
| 110 |
+
"151656": {
|
| 111 |
+
"content": "<|video_pad|>",
|
| 112 |
+
"lstrip": false,
|
| 113 |
+
"normalized": false,
|
| 114 |
+
"rstrip": false,
|
| 115 |
+
"single_word": false,
|
| 116 |
+
"special": true
|
| 117 |
+
},
|
| 118 |
+
"151657": {
|
| 119 |
+
"content": "<tool_call>",
|
| 120 |
+
"lstrip": false,
|
| 121 |
+
"normalized": false,
|
| 122 |
+
"rstrip": false,
|
| 123 |
+
"single_word": false,
|
| 124 |
+
"special": false
|
| 125 |
+
},
|
| 126 |
+
"151658": {
|
| 127 |
+
"content": "</tool_call>",
|
| 128 |
+
"lstrip": false,
|
| 129 |
+
"normalized": false,
|
| 130 |
+
"rstrip": false,
|
| 131 |
+
"single_word": false,
|
| 132 |
+
"special": false
|
| 133 |
+
},
|
| 134 |
+
"151659": {
|
| 135 |
+
"content": "<|fim_prefix|>",
|
| 136 |
+
"lstrip": false,
|
| 137 |
+
"normalized": false,
|
| 138 |
+
"rstrip": false,
|
| 139 |
+
"single_word": false,
|
| 140 |
+
"special": false
|
| 141 |
+
},
|
| 142 |
+
"151660": {
|
| 143 |
+
"content": "<|fim_middle|>",
|
| 144 |
+
"lstrip": false,
|
| 145 |
+
"normalized": false,
|
| 146 |
+
"rstrip": false,
|
| 147 |
+
"single_word": false,
|
| 148 |
+
"special": false
|
| 149 |
+
},
|
| 150 |
+
"151661": {
|
| 151 |
+
"content": "<|fim_suffix|>",
|
| 152 |
+
"lstrip": false,
|
| 153 |
+
"normalized": false,
|
| 154 |
+
"rstrip": false,
|
| 155 |
+
"single_word": false,
|
| 156 |
+
"special": false
|
| 157 |
+
},
|
| 158 |
+
"151662": {
|
| 159 |
+
"content": "<|fim_pad|>",
|
| 160 |
+
"lstrip": false,
|
| 161 |
+
"normalized": false,
|
| 162 |
+
"rstrip": false,
|
| 163 |
+
"single_word": false,
|
| 164 |
+
"special": false
|
| 165 |
+
},
|
| 166 |
+
"151663": {
|
| 167 |
+
"content": "<|repo_name|>",
|
| 168 |
+
"lstrip": false,
|
| 169 |
+
"normalized": false,
|
| 170 |
+
"rstrip": false,
|
| 171 |
+
"single_word": false,
|
| 172 |
+
"special": false
|
| 173 |
+
},
|
| 174 |
+
"151664": {
|
| 175 |
+
"content": "<|file_sep|>",
|
| 176 |
+
"lstrip": false,
|
| 177 |
+
"normalized": false,
|
| 178 |
+
"rstrip": false,
|
| 179 |
+
"single_word": false,
|
| 180 |
+
"special": false
|
| 181 |
+
},
|
| 182 |
+
"151665": {
|
| 183 |
+
"content": "<tool_response>",
|
| 184 |
+
"lstrip": false,
|
| 185 |
+
"normalized": false,
|
| 186 |
+
"rstrip": false,
|
| 187 |
+
"single_word": false,
|
| 188 |
+
"special": false
|
| 189 |
+
},
|
| 190 |
+
"151666": {
|
| 191 |
+
"content": "</tool_response>",
|
| 192 |
+
"lstrip": false,
|
| 193 |
+
"normalized": false,
|
| 194 |
+
"rstrip": false,
|
| 195 |
+
"single_word": false,
|
| 196 |
+
"special": false
|
| 197 |
+
},
|
| 198 |
+
"151667": {
|
| 199 |
+
"content": "<think>",
|
| 200 |
+
"lstrip": false,
|
| 201 |
+
"normalized": false,
|
| 202 |
+
"rstrip": false,
|
| 203 |
+
"single_word": false,
|
| 204 |
+
"special": false
|
| 205 |
+
},
|
| 206 |
+
"151668": {
|
| 207 |
+
"content": "</think>",
|
| 208 |
+
"lstrip": false,
|
| 209 |
+
"normalized": false,
|
| 210 |
+
"rstrip": false,
|
| 211 |
+
"single_word": false,
|
| 212 |
+
"special": false
|
| 213 |
+
},
|
| 214 |
+
"151669": {
|
| 215 |
+
"content": "<img>",
|
| 216 |
+
"lstrip": false,
|
| 217 |
+
"normalized": false,
|
| 218 |
+
"rstrip": false,
|
| 219 |
+
"single_word": false,
|
| 220 |
+
"special": true
|
| 221 |
+
},
|
| 222 |
+
"151670": {
|
| 223 |
+
"content": "</img>",
|
| 224 |
+
"lstrip": false,
|
| 225 |
+
"normalized": false,
|
| 226 |
+
"rstrip": false,
|
| 227 |
+
"single_word": false,
|
| 228 |
+
"special": true
|
| 229 |
+
},
|
| 230 |
+
"151671": {
|
| 231 |
+
"content": "<IMG_CONTEXT>",
|
| 232 |
+
"lstrip": false,
|
| 233 |
+
"normalized": false,
|
| 234 |
+
"rstrip": false,
|
| 235 |
+
"single_word": false,
|
| 236 |
+
"special": true
|
| 237 |
+
},
|
| 238 |
+
"151672": {
|
| 239 |
+
"content": "<quad>",
|
| 240 |
+
"lstrip": false,
|
| 241 |
+
"normalized": false,
|
| 242 |
+
"rstrip": false,
|
| 243 |
+
"single_word": false,
|
| 244 |
+
"special": true
|
| 245 |
+
},
|
| 246 |
+
"151673": {
|
| 247 |
+
"content": "</quad>",
|
| 248 |
+
"lstrip": false,
|
| 249 |
+
"normalized": false,
|
| 250 |
+
"rstrip": false,
|
| 251 |
+
"single_word": false,
|
| 252 |
+
"special": true
|
| 253 |
+
},
|
| 254 |
+
"151674": {
|
| 255 |
+
"content": "<ref>",
|
| 256 |
+
"lstrip": false,
|
| 257 |
+
"normalized": false,
|
| 258 |
+
"rstrip": false,
|
| 259 |
+
"single_word": false,
|
| 260 |
+
"special": true
|
| 261 |
+
},
|
| 262 |
+
"151675": {
|
| 263 |
+
"content": "</ref>",
|
| 264 |
+
"lstrip": false,
|
| 265 |
+
"normalized": false,
|
| 266 |
+
"rstrip": false,
|
| 267 |
+
"single_word": false,
|
| 268 |
+
"special": true
|
| 269 |
+
},
|
| 270 |
+
"151676": {
|
| 271 |
+
"content": "<box>",
|
| 272 |
+
"lstrip": false,
|
| 273 |
+
"normalized": false,
|
| 274 |
+
"rstrip": false,
|
| 275 |
+
"single_word": false,
|
| 276 |
+
"special": true
|
| 277 |
+
},
|
| 278 |
+
"151677": {
|
| 279 |
+
"content": "</box>",
|
| 280 |
+
"lstrip": false,
|
| 281 |
+
"normalized": false,
|
| 282 |
+
"rstrip": false,
|
| 283 |
+
"single_word": false,
|
| 284 |
+
"special": true
|
| 285 |
+
}
|
| 286 |
+
},
|
| 287 |
+
"additional_special_tokens": [
|
| 288 |
+
"<|im_start|>",
|
| 289 |
+
"<|im_end|>",
|
| 290 |
+
"<|object_ref_start|>",
|
| 291 |
+
"<|object_ref_end|>",
|
| 292 |
+
"<|box_start|>",
|
| 293 |
+
"<|box_end|>",
|
| 294 |
+
"<|quad_start|>",
|
| 295 |
+
"<|quad_end|>",
|
| 296 |
+
"<|vision_start|>",
|
| 297 |
+
"<|vision_end|>",
|
| 298 |
+
"<|vision_pad|>",
|
| 299 |
+
"<|image_pad|>",
|
| 300 |
+
"<|video_pad|>",
|
| 301 |
+
"<img>",
|
| 302 |
+
"</img>",
|
| 303 |
+
"<IMG_CONTEXT>",
|
| 304 |
+
"<quad>",
|
| 305 |
+
"</quad>",
|
| 306 |
+
"<ref>",
|
| 307 |
+
"</ref>",
|
| 308 |
+
"<box>",
|
| 309 |
+
"</box>"
|
| 310 |
+
],
|
| 311 |
+
"bos_token": null,
|
| 312 |
+
"clean_up_tokenization_spaces": false,
|
| 313 |
+
"eos_token": "<|im_end|>",
|
| 314 |
+
"errors": "replace",
|
| 315 |
+
"extra_special_tokens": {},
|
| 316 |
+
"model_max_length": 14588,
|
| 317 |
+
"pad_token": "<|endoftext|>",
|
| 318 |
+
"processor_class": "InternVLProcessor",
|
| 319 |
+
"split_special_tokens": false,
|
| 320 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 321 |
+
"unk_token": null
|
| 322 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,934 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.3472222222222222,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 100,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"avg_label": 0.45932112634181976,
|
| 14 |
+
"epoch": 0.003472222222222222,
|
| 15 |
+
"grad_norm": 0.515625,
|
| 16 |
+
"learning_rate": 0.0,
|
| 17 |
+
"loss": 0.02412083954550326,
|
| 18 |
+
"num_pairs": 28.0,
|
| 19 |
+
"step": 1
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"avg_label": 0.4166666865348816,
|
| 23 |
+
"epoch": 0.006944444444444444,
|
| 24 |
+
"grad_norm": 11.625,
|
| 25 |
+
"learning_rate": 6.896551724137931e-07,
|
| 26 |
+
"loss": 0.40197963267564774,
|
| 27 |
+
"num_pairs": 21.5,
|
| 28 |
+
"step": 2
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"avg_label": 0.4724903553724289,
|
| 32 |
+
"epoch": 0.010416666666666666,
|
| 33 |
+
"grad_norm": 13.25,
|
| 34 |
+
"learning_rate": 1.3793103448275862e-06,
|
| 35 |
+
"loss": 0.5004251897335052,
|
| 36 |
+
"num_pairs": 28.0,
|
| 37 |
+
"step": 3
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"avg_label": 0.44384056329727173,
|
| 41 |
+
"epoch": 0.013888888888888888,
|
| 42 |
+
"grad_norm": 5.875,
|
| 43 |
+
"learning_rate": 2.0689655172413796e-06,
|
| 44 |
+
"loss": 0.23160110414028168,
|
| 45 |
+
"num_pairs": 27.5,
|
| 46 |
+
"step": 4
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"avg_label": 0.4630681872367859,
|
| 50 |
+
"epoch": 0.017361111111111112,
|
| 51 |
+
"grad_norm": 8.9375,
|
| 52 |
+
"learning_rate": 2.7586206896551725e-06,
|
| 53 |
+
"loss": 0.4468539021909237,
|
| 54 |
+
"num_pairs": 24.0,
|
| 55 |
+
"step": 5
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"avg_label": 0.45495015382766724,
|
| 59 |
+
"epoch": 0.020833333333333332,
|
| 60 |
+
"grad_norm": 1.34375,
|
| 61 |
+
"learning_rate": 3.448275862068966e-06,
|
| 62 |
+
"loss": 0.017617323901504278,
|
| 63 |
+
"num_pairs": 28.0,
|
| 64 |
+
"step": 6
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"avg_label": 0.4732142984867096,
|
| 68 |
+
"epoch": 0.024305555555555556,
|
| 69 |
+
"grad_norm": 11.5625,
|
| 70 |
+
"learning_rate": 4.137931034482759e-06,
|
| 71 |
+
"loss": 0.4206290766596794,
|
| 72 |
+
"num_pairs": 24.0,
|
| 73 |
+
"step": 7
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"avg_label": 0.4736842066049576,
|
| 77 |
+
"epoch": 0.027777777777777776,
|
| 78 |
+
"grad_norm": 0.1484375,
|
| 79 |
+
"learning_rate": 4.8275862068965525e-06,
|
| 80 |
+
"loss": 0.004913174081593752,
|
| 81 |
+
"num_pairs": 27.0,
|
| 82 |
+
"step": 8
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"avg_label": 0.45513392984867096,
|
| 86 |
+
"epoch": 0.03125,
|
| 87 |
+
"grad_norm": 7.84375,
|
| 88 |
+
"learning_rate": 5.517241379310345e-06,
|
| 89 |
+
"loss": 0.40890943817794323,
|
| 90 |
+
"num_pairs": 28.0,
|
| 91 |
+
"step": 9
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"avg_label": 0.4797794222831726,
|
| 95 |
+
"epoch": 0.034722222222222224,
|
| 96 |
+
"grad_norm": 4.21875,
|
| 97 |
+
"learning_rate": 6.206896551724138e-06,
|
| 98 |
+
"loss": 0.07930763380136341,
|
| 99 |
+
"num_pairs": 28.0,
|
| 100 |
+
"step": 10
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"avg_label": 0.4505208283662796,
|
| 104 |
+
"epoch": 0.03819444444444445,
|
| 105 |
+
"grad_norm": 18.0,
|
| 106 |
+
"learning_rate": 6.896551724137932e-06,
|
| 107 |
+
"loss": 0.8742586374282837,
|
| 108 |
+
"num_pairs": 28.0,
|
| 109 |
+
"step": 11
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"avg_label": 0.42196430265903473,
|
| 113 |
+
"epoch": 0.041666666666666664,
|
| 114 |
+
"grad_norm": 17.625,
|
| 115 |
+
"learning_rate": 7.586206896551724e-06,
|
| 116 |
+
"loss": 0.8603673875331879,
|
| 117 |
+
"num_pairs": 23.5,
|
| 118 |
+
"step": 12
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"avg_label": 0.44431088864803314,
|
| 122 |
+
"epoch": 0.04513888888888889,
|
| 123 |
+
"grad_norm": 1.3203125,
|
| 124 |
+
"learning_rate": 8.275862068965518e-06,
|
| 125 |
+
"loss": 0.028431319631636143,
|
| 126 |
+
"num_pairs": 28.0,
|
| 127 |
+
"step": 13
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"avg_label": 0.3893229216337204,
|
| 131 |
+
"epoch": 0.04861111111111111,
|
| 132 |
+
"grad_norm": 10.0,
|
| 133 |
+
"learning_rate": 8.965517241379312e-06,
|
| 134 |
+
"loss": 0.251303069293499,
|
| 135 |
+
"num_pairs": 23.5,
|
| 136 |
+
"step": 14
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"avg_label": 0.43348929286003113,
|
| 140 |
+
"epoch": 0.052083333333333336,
|
| 141 |
+
"grad_norm": 3.796875,
|
| 142 |
+
"learning_rate": 9.655172413793105e-06,
|
| 143 |
+
"loss": 0.11253293231129646,
|
| 144 |
+
"num_pairs": 27.0,
|
| 145 |
+
"step": 15
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"avg_label": 0.439510241150856,
|
| 149 |
+
"epoch": 0.05555555555555555,
|
| 150 |
+
"grad_norm": 4.21875,
|
| 151 |
+
"learning_rate": 1.0344827586206898e-05,
|
| 152 |
+
"loss": 0.16489703953266144,
|
| 153 |
+
"num_pairs": 28.0,
|
| 154 |
+
"step": 16
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"avg_label": 0.4412849396467209,
|
| 158 |
+
"epoch": 0.059027777777777776,
|
| 159 |
+
"grad_norm": 4.84375,
|
| 160 |
+
"learning_rate": 1.103448275862069e-05,
|
| 161 |
+
"loss": 0.17035143449902534,
|
| 162 |
+
"num_pairs": 28.0,
|
| 163 |
+
"step": 17
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"avg_label": 0.4661668390035629,
|
| 167 |
+
"epoch": 0.0625,
|
| 168 |
+
"grad_norm": 3.1875,
|
| 169 |
+
"learning_rate": 1.1724137931034483e-05,
|
| 170 |
+
"loss": 0.10208164528012276,
|
| 171 |
+
"num_pairs": 28.0,
|
| 172 |
+
"step": 18
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"avg_label": 0.45625001192092896,
|
| 176 |
+
"epoch": 0.06597222222222222,
|
| 177 |
+
"grad_norm": 13.625,
|
| 178 |
+
"learning_rate": 1.2413793103448277e-05,
|
| 179 |
+
"loss": 0.5987264770083129,
|
| 180 |
+
"num_pairs": 28.0,
|
| 181 |
+
"step": 19
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"avg_label": 0.4458743929862976,
|
| 185 |
+
"epoch": 0.06944444444444445,
|
| 186 |
+
"grad_norm": 0.59375,
|
| 187 |
+
"learning_rate": 1.310344827586207e-05,
|
| 188 |
+
"loss": 0.03994415677152574,
|
| 189 |
+
"num_pairs": 28.0,
|
| 190 |
+
"step": 20
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"avg_label": 0.4499442130327225,
|
| 194 |
+
"epoch": 0.07291666666666667,
|
| 195 |
+
"grad_norm": 2.078125,
|
| 196 |
+
"learning_rate": 1.3793103448275863e-05,
|
| 197 |
+
"loss": 0.08950073830783367,
|
| 198 |
+
"num_pairs": 28.0,
|
| 199 |
+
"step": 21
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"avg_label": 0.4587053656578064,
|
| 203 |
+
"epoch": 0.0763888888888889,
|
| 204 |
+
"grad_norm": 9.0625,
|
| 205 |
+
"learning_rate": 1.4482758620689657e-05,
|
| 206 |
+
"loss": 0.5212011188268661,
|
| 207 |
+
"num_pairs": 28.0,
|
| 208 |
+
"step": 22
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"avg_label": 0.4502655118703842,
|
| 212 |
+
"epoch": 0.0798611111111111,
|
| 213 |
+
"grad_norm": 5.03125,
|
| 214 |
+
"learning_rate": 1.5172413793103448e-05,
|
| 215 |
+
"loss": 0.08295552420895547,
|
| 216 |
+
"num_pairs": 28.0,
|
| 217 |
+
"step": 23
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"avg_label": 0.4362068921327591,
|
| 221 |
+
"epoch": 0.08333333333333333,
|
| 222 |
+
"grad_norm": 1.65625,
|
| 223 |
+
"learning_rate": 1.586206896551724e-05,
|
| 224 |
+
"loss": 0.07221807166934013,
|
| 225 |
+
"num_pairs": 27.0,
|
| 226 |
+
"step": 24
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"avg_label": 0.4532342702150345,
|
| 230 |
+
"epoch": 0.08680555555555555,
|
| 231 |
+
"grad_norm": 1.140625,
|
| 232 |
+
"learning_rate": 1.6551724137931037e-05,
|
| 233 |
+
"loss": 0.03893738705664873,
|
| 234 |
+
"num_pairs": 28.0,
|
| 235 |
+
"step": 25
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"avg_label": 0.4947916716337204,
|
| 239 |
+
"epoch": 0.09027777777777778,
|
| 240 |
+
"grad_norm": 6.28125,
|
| 241 |
+
"learning_rate": 1.7241379310344828e-05,
|
| 242 |
+
"loss": 0.13335712999105453,
|
| 243 |
+
"num_pairs": 23.5,
|
| 244 |
+
"step": 26
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"avg_label": 0.3125,
|
| 248 |
+
"epoch": 0.09375,
|
| 249 |
+
"grad_norm": 13.75,
|
| 250 |
+
"learning_rate": 1.7931034482758623e-05,
|
| 251 |
+
"loss": 0.23121396452188492,
|
| 252 |
+
"num_pairs": 17.5,
|
| 253 |
+
"step": 27
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"avg_label": 0.4624594449996948,
|
| 257 |
+
"epoch": 0.09722222222222222,
|
| 258 |
+
"grad_norm": 11.75,
|
| 259 |
+
"learning_rate": 1.8620689655172415e-05,
|
| 260 |
+
"loss": 0.4965571314096451,
|
| 261 |
+
"num_pairs": 28.0,
|
| 262 |
+
"step": 28
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"avg_label": 0.46101999282836914,
|
| 266 |
+
"epoch": 0.10069444444444445,
|
| 267 |
+
"grad_norm": 3.578125,
|
| 268 |
+
"learning_rate": 1.931034482758621e-05,
|
| 269 |
+
"loss": 0.16813608072698116,
|
| 270 |
+
"num_pairs": 28.0,
|
| 271 |
+
"step": 29
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"avg_label": 0.4742646962404251,
|
| 275 |
+
"epoch": 0.10416666666666667,
|
| 276 |
+
"grad_norm": 3.171875,
|
| 277 |
+
"learning_rate": 2e-05,
|
| 278 |
+
"loss": 0.1642257682979107,
|
| 279 |
+
"num_pairs": 22.0,
|
| 280 |
+
"step": 30
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"avg_label": 0.45412661135196686,
|
| 284 |
+
"epoch": 0.1076388888888889,
|
| 285 |
+
"grad_norm": 8.4375,
|
| 286 |
+
"learning_rate": 1.9999835072185805e-05,
|
| 287 |
+
"loss": 0.24545209854841232,
|
| 288 |
+
"num_pairs": 28.0,
|
| 289 |
+
"step": 31
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"avg_label": 0.45036764442920685,
|
| 293 |
+
"epoch": 0.1111111111111111,
|
| 294 |
+
"grad_norm": 3.234375,
|
| 295 |
+
"learning_rate": 1.999934029418346e-05,
|
| 296 |
+
"loss": 0.14530762657523155,
|
| 297 |
+
"num_pairs": 26.0,
|
| 298 |
+
"step": 32
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"avg_label": 0.4388917088508606,
|
| 302 |
+
"epoch": 0.11458333333333333,
|
| 303 |
+
"grad_norm": 2.890625,
|
| 304 |
+
"learning_rate": 1.9998515682313485e-05,
|
| 305 |
+
"loss": 0.10479701962321997,
|
| 306 |
+
"num_pairs": 28.0,
|
| 307 |
+
"step": 33
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"avg_label": 0.43915343284606934,
|
| 311 |
+
"epoch": 0.11805555555555555,
|
| 312 |
+
"grad_norm": 7.875,
|
| 313 |
+
"learning_rate": 1.999736126377618e-05,
|
| 314 |
+
"loss": 0.3016491085290909,
|
| 315 |
+
"num_pairs": 28.0,
|
| 316 |
+
"step": 34
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"avg_label": 0.4497767984867096,
|
| 320 |
+
"epoch": 0.12152777777777778,
|
| 321 |
+
"grad_norm": 3.96875,
|
| 322 |
+
"learning_rate": 1.999587707665068e-05,
|
| 323 |
+
"loss": 0.15345897153019905,
|
| 324 |
+
"num_pairs": 28.0,
|
| 325 |
+
"step": 35
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"avg_label": 0.4572916626930237,
|
| 329 |
+
"epoch": 0.125,
|
| 330 |
+
"grad_norm": 1.3515625,
|
| 331 |
+
"learning_rate": 1.999406316989374e-05,
|
| 332 |
+
"loss": 0.04903313983231783,
|
| 333 |
+
"num_pairs": 28.0,
|
| 334 |
+
"step": 36
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"avg_label": 0.45703125,
|
| 338 |
+
"epoch": 0.1284722222222222,
|
| 339 |
+
"grad_norm": 8.6875,
|
| 340 |
+
"learning_rate": 1.9991919603338088e-05,
|
| 341 |
+
"loss": 0.36730772256851196,
|
| 342 |
+
"num_pairs": 28.0,
|
| 343 |
+
"step": 37
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"avg_label": 0.4379105120897293,
|
| 347 |
+
"epoch": 0.13194444444444445,
|
| 348 |
+
"grad_norm": 6.25,
|
| 349 |
+
"learning_rate": 1.998944644769048e-05,
|
| 350 |
+
"loss": 0.2889542682096362,
|
| 351 |
+
"num_pairs": 27.5,
|
| 352 |
+
"step": 38
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"avg_label": 0.4781250059604645,
|
| 356 |
+
"epoch": 0.13541666666666666,
|
| 357 |
+
"grad_norm": 10.0625,
|
| 358 |
+
"learning_rate": 1.9986643784529346e-05,
|
| 359 |
+
"loss": 0.416560098528862,
|
| 360 |
+
"num_pairs": 24.5,
|
| 361 |
+
"step": 39
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"avg_label": 0.4678819328546524,
|
| 365 |
+
"epoch": 0.1388888888888889,
|
| 366 |
+
"grad_norm": 4.96875,
|
| 367 |
+
"learning_rate": 1.9983511706302102e-05,
|
| 368 |
+
"loss": 0.1765335127711296,
|
| 369 |
+
"num_pairs": 28.0,
|
| 370 |
+
"step": 40
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"avg_label": 0.4556451737880707,
|
| 374 |
+
"epoch": 0.1423611111111111,
|
| 375 |
+
"grad_norm": 3.171875,
|
| 376 |
+
"learning_rate": 1.9980050316322118e-05,
|
| 377 |
+
"loss": 0.15112879872322083,
|
| 378 |
+
"num_pairs": 27.5,
|
| 379 |
+
"step": 41
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"avg_label": 0.4666379243135452,
|
| 383 |
+
"epoch": 0.14583333333333334,
|
| 384 |
+
"grad_norm": 10.875,
|
| 385 |
+
"learning_rate": 1.997625972876529e-05,
|
| 386 |
+
"loss": 0.5556365996599197,
|
| 387 |
+
"num_pairs": 28.0,
|
| 388 |
+
"step": 42
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"avg_label": 0.45076756179332733,
|
| 392 |
+
"epoch": 0.14930555555555555,
|
| 393 |
+
"grad_norm": 2.40625,
|
| 394 |
+
"learning_rate": 1.997214006866628e-05,
|
| 395 |
+
"loss": 0.09973586304113269,
|
| 396 |
+
"num_pairs": 28.0,
|
| 397 |
+
"step": 43
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"avg_label": 0.4405048042535782,
|
| 401 |
+
"epoch": 0.1527777777777778,
|
| 402 |
+
"grad_norm": 8.9375,
|
| 403 |
+
"learning_rate": 1.9967691471914392e-05,
|
| 404 |
+
"loss": 0.3806912750005722,
|
| 405 |
+
"num_pairs": 28.0,
|
| 406 |
+
"step": 44
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"avg_label": 0.4602416008710861,
|
| 410 |
+
"epoch": 0.15625,
|
| 411 |
+
"grad_norm": 4.8125,
|
| 412 |
+
"learning_rate": 1.99629140852491e-05,
|
| 413 |
+
"loss": 0.14190024323761463,
|
| 414 |
+
"num_pairs": 28.0,
|
| 415 |
+
"step": 45
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"avg_label": 0.44902269542217255,
|
| 419 |
+
"epoch": 0.1597222222222222,
|
| 420 |
+
"grad_norm": 1.890625,
|
| 421 |
+
"learning_rate": 1.9957808066255187e-05,
|
| 422 |
+
"loss": 0.07964974269270897,
|
| 423 |
+
"num_pairs": 28.0,
|
| 424 |
+
"step": 46
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"avg_label": 0.44250571727752686,
|
| 428 |
+
"epoch": 0.16319444444444445,
|
| 429 |
+
"grad_norm": 7.5625,
|
| 430 |
+
"learning_rate": 1.9952373583357566e-05,
|
| 431 |
+
"loss": 0.3196651563048363,
|
| 432 |
+
"num_pairs": 27.5,
|
| 433 |
+
"step": 47
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"avg_label": 0.45804399251937866,
|
| 437 |
+
"epoch": 0.16666666666666666,
|
| 438 |
+
"grad_norm": 3.03125,
|
| 439 |
+
"learning_rate": 1.994661081581571e-05,
|
| 440 |
+
"loss": 0.1772424913942814,
|
| 441 |
+
"num_pairs": 28.0,
|
| 442 |
+
"step": 48
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"avg_label": 0.44343800842761993,
|
| 446 |
+
"epoch": 0.1701388888888889,
|
| 447 |
+
"grad_norm": 0.7734375,
|
| 448 |
+
"learning_rate": 1.9940519953717762e-05,
|
| 449 |
+
"loss": 0.03773397207260132,
|
| 450 |
+
"num_pairs": 27.5,
|
| 451 |
+
"step": 49
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"avg_label": 0.4277420938014984,
|
| 455 |
+
"epoch": 0.1736111111111111,
|
| 456 |
+
"grad_norm": 0.7578125,
|
| 457 |
+
"learning_rate": 1.993410119797422e-05,
|
| 458 |
+
"loss": 0.017126482212916017,
|
| 459 |
+
"num_pairs": 27.5,
|
| 460 |
+
"step": 50
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"avg_label": 0.4761904776096344,
|
| 464 |
+
"epoch": 0.17708333333333334,
|
| 465 |
+
"grad_norm": 2.265625,
|
| 466 |
+
"learning_rate": 1.9927354760311365e-05,
|
| 467 |
+
"loss": 0.0958416610956192,
|
| 468 |
+
"num_pairs": 24.0,
|
| 469 |
+
"step": 51
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"avg_label": 0.45110294222831726,
|
| 473 |
+
"epoch": 0.18055555555555555,
|
| 474 |
+
"grad_norm": 3.859375,
|
| 475 |
+
"learning_rate": 1.992028086326424e-05,
|
| 476 |
+
"loss": 0.176346430554986,
|
| 477 |
+
"num_pairs": 28.0,
|
| 478 |
+
"step": 52
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"avg_label": 0.4557291716337204,
|
| 482 |
+
"epoch": 0.1840277777777778,
|
| 483 |
+
"grad_norm": 1.484375,
|
| 484 |
+
"learning_rate": 1.991287974016932e-05,
|
| 485 |
+
"loss": 0.06965811923146248,
|
| 486 |
+
"num_pairs": 28.0,
|
| 487 |
+
"step": 53
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"avg_label": 0.441498339176178,
|
| 491 |
+
"epoch": 0.1875,
|
| 492 |
+
"grad_norm": 1.1328125,
|
| 493 |
+
"learning_rate": 1.9905151635156813e-05,
|
| 494 |
+
"loss": 0.04886321909725666,
|
| 495 |
+
"num_pairs": 28.0,
|
| 496 |
+
"step": 54
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"avg_label": 0.4579603523015976,
|
| 500 |
+
"epoch": 0.1909722222222222,
|
| 501 |
+
"grad_norm": 0.79296875,
|
| 502 |
+
"learning_rate": 1.9897096803142616e-05,
|
| 503 |
+
"loss": 0.02616998180747032,
|
| 504 |
+
"num_pairs": 28.0,
|
| 505 |
+
"step": 55
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"avg_label": 0.45521390438079834,
|
| 509 |
+
"epoch": 0.19444444444444445,
|
| 510 |
+
"grad_norm": 0.48046875,
|
| 511 |
+
"learning_rate": 1.988871550981989e-05,
|
| 512 |
+
"loss": 0.04496368020772934,
|
| 513 |
+
"num_pairs": 28.0,
|
| 514 |
+
"step": 56
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"avg_label": 0.46743176877498627,
|
| 518 |
+
"epoch": 0.19791666666666666,
|
| 519 |
+
"grad_norm": 1.953125,
|
| 520 |
+
"learning_rate": 1.988000803165032e-05,
|
| 521 |
+
"loss": 0.08031387068331242,
|
| 522 |
+
"num_pairs": 27.0,
|
| 523 |
+
"step": 57
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"avg_label": 0.39706501364707947,
|
| 527 |
+
"epoch": 0.2013888888888889,
|
| 528 |
+
"grad_norm": 0.765625,
|
| 529 |
+
"learning_rate": 1.9870974655854974e-05,
|
| 530 |
+
"loss": 0.05276821367442608,
|
| 531 |
+
"num_pairs": 28.0,
|
| 532 |
+
"step": 58
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"avg_label": 0.4635416716337204,
|
| 536 |
+
"epoch": 0.2048611111111111,
|
| 537 |
+
"grad_norm": 0.345703125,
|
| 538 |
+
"learning_rate": 1.9861615680404833e-05,
|
| 539 |
+
"loss": 0.02046751804300584,
|
| 540 |
+
"num_pairs": 28.0,
|
| 541 |
+
"step": 59
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"avg_label": 0.4392857253551483,
|
| 545 |
+
"epoch": 0.20833333333333334,
|
| 546 |
+
"grad_norm": 1.59375,
|
| 547 |
+
"learning_rate": 1.985193141401097e-05,
|
| 548 |
+
"loss": 0.03745671547949314,
|
| 549 |
+
"num_pairs": 28.0,
|
| 550 |
+
"step": 60
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"avg_label": 0.430803582072258,
|
| 554 |
+
"epoch": 0.21180555555555555,
|
| 555 |
+
"grad_norm": 2.34375,
|
| 556 |
+
"learning_rate": 1.9841922176114366e-05,
|
| 557 |
+
"loss": 0.04785962332971394,
|
| 558 |
+
"num_pairs": 21.5,
|
| 559 |
+
"step": 61
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"avg_label": 0.4626736044883728,
|
| 563 |
+
"epoch": 0.2152777777777778,
|
| 564 |
+
"grad_norm": 0.61328125,
|
| 565 |
+
"learning_rate": 1.9831588296875367e-05,
|
| 566 |
+
"loss": 0.021605145651847124,
|
| 567 |
+
"num_pairs": 28.0,
|
| 568 |
+
"step": 62
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"avg_label": 0.4780505895614624,
|
| 572 |
+
"epoch": 0.21875,
|
| 573 |
+
"grad_norm": 1.609375,
|
| 574 |
+
"learning_rate": 1.982093011716279e-05,
|
| 575 |
+
"loss": 0.08873376995325089,
|
| 576 |
+
"num_pairs": 28.0,
|
| 577 |
+
"step": 63
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"avg_label": 0.4489087462425232,
|
| 581 |
+
"epoch": 0.2222222222222222,
|
| 582 |
+
"grad_norm": 1.65625,
|
| 583 |
+
"learning_rate": 1.9809947988542696e-05,
|
| 584 |
+
"loss": 0.0287565803155303,
|
| 585 |
+
"num_pairs": 27.5,
|
| 586 |
+
"step": 64
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"avg_label": 0.4442349076271057,
|
| 590 |
+
"epoch": 0.22569444444444445,
|
| 591 |
+
"grad_norm": 0.82421875,
|
| 592 |
+
"learning_rate": 1.979864227326678e-05,
|
| 593 |
+
"loss": 0.041462352965027094,
|
| 594 |
+
"num_pairs": 28.0,
|
| 595 |
+
"step": 65
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"avg_label": 0.5274767875671387,
|
| 599 |
+
"epoch": 0.22916666666666666,
|
| 600 |
+
"grad_norm": 0.26171875,
|
| 601 |
+
"learning_rate": 1.9787013344260422e-05,
|
| 602 |
+
"loss": 0.006527273333631456,
|
| 603 |
+
"num_pairs": 26.0,
|
| 604 |
+
"step": 66
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"avg_label": 0.45734797418117523,
|
| 608 |
+
"epoch": 0.2326388888888889,
|
| 609 |
+
"grad_norm": 7.21875,
|
| 610 |
+
"learning_rate": 1.9775061585110387e-05,
|
| 611 |
+
"loss": 0.33710330724716187,
|
| 612 |
+
"num_pairs": 28.0,
|
| 613 |
+
"step": 67
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"avg_label": 0.44953545928001404,
|
| 617 |
+
"epoch": 0.2361111111111111,
|
| 618 |
+
"grad_norm": 1.1796875,
|
| 619 |
+
"learning_rate": 1.976278739005218e-05,
|
| 620 |
+
"loss": 0.05175479780882597,
|
| 621 |
+
"num_pairs": 28.0,
|
| 622 |
+
"step": 68
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"avg_label": 0.4675245136022568,
|
| 626 |
+
"epoch": 0.23958333333333334,
|
| 627 |
+
"grad_norm": 3.859375,
|
| 628 |
+
"learning_rate": 1.9750191163957042e-05,
|
| 629 |
+
"loss": 0.15232571214437485,
|
| 630 |
+
"num_pairs": 28.0,
|
| 631 |
+
"step": 69
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"avg_label": 0.45781250298023224,
|
| 635 |
+
"epoch": 0.24305555555555555,
|
| 636 |
+
"grad_norm": 1.7890625,
|
| 637 |
+
"learning_rate": 1.9737273322318565e-05,
|
| 638 |
+
"loss": 0.09527681209146976,
|
| 639 |
+
"num_pairs": 28.0,
|
| 640 |
+
"step": 70
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"avg_label": 0.4518994987010956,
|
| 644 |
+
"epoch": 0.2465277777777778,
|
| 645 |
+
"grad_norm": 0.703125,
|
| 646 |
+
"learning_rate": 1.972403429123904e-05,
|
| 647 |
+
"loss": 0.035001321870367974,
|
| 648 |
+
"num_pairs": 28.0,
|
| 649 |
+
"step": 71
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"avg_label": 0.4765625,
|
| 653 |
+
"epoch": 0.25,
|
| 654 |
+
"grad_norm": 0.75,
|
| 655 |
+
"learning_rate": 1.971047450741535e-05,
|
| 656 |
+
"loss": 0.026095230132341385,
|
| 657 |
+
"num_pairs": 22.0,
|
| 658 |
+
"step": 72
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"avg_label": 0.5010775923728943,
|
| 662 |
+
"epoch": 0.2534722222222222,
|
| 663 |
+
"grad_norm": 0.58984375,
|
| 664 |
+
"learning_rate": 1.9696594418124598e-05,
|
| 665 |
+
"loss": 0.028016670839861035,
|
| 666 |
+
"num_pairs": 23.5,
|
| 667 |
+
"step": 73
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"avg_label": 0.6005434691905975,
|
| 671 |
+
"epoch": 0.2569444444444444,
|
| 672 |
+
"grad_norm": 0.62109375,
|
| 673 |
+
"learning_rate": 1.9682394481209338e-05,
|
| 674 |
+
"loss": 0.017244269140064716,
|
| 675 |
+
"num_pairs": 20.0,
|
| 676 |
+
"step": 74
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"avg_label": 0.44895362854003906,
|
| 680 |
+
"epoch": 0.2604166666666667,
|
| 681 |
+
"grad_norm": 2.65625,
|
| 682 |
+
"learning_rate": 1.966787516506249e-05,
|
| 683 |
+
"loss": 0.13350838515907526,
|
| 684 |
+
"num_pairs": 27.5,
|
| 685 |
+
"step": 75
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"avg_label": 0.44657258689403534,
|
| 689 |
+
"epoch": 0.2638888888888889,
|
| 690 |
+
"grad_norm": 2.6875,
|
| 691 |
+
"learning_rate": 1.9653036948611864e-05,
|
| 692 |
+
"loss": 0.13598253019154072,
|
| 693 |
+
"num_pairs": 24.5,
|
| 694 |
+
"step": 76
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"avg_label": 0.4062500149011612,
|
| 698 |
+
"epoch": 0.2673611111111111,
|
| 699 |
+
"grad_norm": 1.6328125,
|
| 700 |
+
"learning_rate": 1.9637880321304387e-05,
|
| 701 |
+
"loss": 0.09597307164222002,
|
| 702 |
+
"num_pairs": 21.5,
|
| 703 |
+
"step": 77
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"avg_label": 0.43358150124549866,
|
| 707 |
+
"epoch": 0.2708333333333333,
|
| 708 |
+
"grad_norm": 1.5859375,
|
| 709 |
+
"learning_rate": 1.962240578308993e-05,
|
| 710 |
+
"loss": 0.02830888982862234,
|
| 711 |
+
"num_pairs": 28.0,
|
| 712 |
+
"step": 78
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"avg_label": 0.4296875,
|
| 716 |
+
"epoch": 0.2743055555555556,
|
| 717 |
+
"grad_norm": 0.5703125,
|
| 718 |
+
"learning_rate": 1.9606613844404853e-05,
|
| 719 |
+
"loss": 0.0342620718292892,
|
| 720 |
+
"num_pairs": 27.5,
|
| 721 |
+
"step": 79
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"avg_label": 0.44969919323921204,
|
| 725 |
+
"epoch": 0.2777777777777778,
|
| 726 |
+
"grad_norm": 1.140625,
|
| 727 |
+
"learning_rate": 1.9590505026155146e-05,
|
| 728 |
+
"loss": 0.047657732968218625,
|
| 729 |
+
"num_pairs": 28.0,
|
| 730 |
+
"step": 80
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"avg_label": 0.44572368264198303,
|
| 734 |
+
"epoch": 0.28125,
|
| 735 |
+
"grad_norm": 0.98828125,
|
| 736 |
+
"learning_rate": 1.9574079859699236e-05,
|
| 737 |
+
"loss": 0.048789044842123985,
|
| 738 |
+
"num_pairs": 28.0,
|
| 739 |
+
"step": 81
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"avg_label": 0.39816810190677643,
|
| 743 |
+
"epoch": 0.2847222222222222,
|
| 744 |
+
"grad_norm": 1.140625,
|
| 745 |
+
"learning_rate": 1.955733888683049e-05,
|
| 746 |
+
"loss": 0.053058773279190063,
|
| 747 |
+
"num_pairs": 27.5,
|
| 748 |
+
"step": 82
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"avg_label": 0.43917112052440643,
|
| 752 |
+
"epoch": 0.2881944444444444,
|
| 753 |
+
"grad_norm": 0.91015625,
|
| 754 |
+
"learning_rate": 1.9540282659759317e-05,
|
| 755 |
+
"loss": 0.05478274542838335,
|
| 756 |
+
"num_pairs": 27.5,
|
| 757 |
+
"step": 83
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"avg_label": 0.4709821492433548,
|
| 761 |
+
"epoch": 0.2916666666666667,
|
| 762 |
+
"grad_norm": 0.228515625,
|
| 763 |
+
"learning_rate": 1.9522911741094966e-05,
|
| 764 |
+
"loss": 0.011658322997391224,
|
| 765 |
+
"num_pairs": 28.0,
|
| 766 |
+
"step": 84
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"avg_label": 0.4544480890035629,
|
| 770 |
+
"epoch": 0.2951388888888889,
|
| 771 |
+
"grad_norm": 11.3125,
|
| 772 |
+
"learning_rate": 1.9505226703826973e-05,
|
| 773 |
+
"loss": 0.523826252669096,
|
| 774 |
+
"num_pairs": 28.0,
|
| 775 |
+
"step": 85
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"avg_label": 0.4711538553237915,
|
| 779 |
+
"epoch": 0.2986111111111111,
|
| 780 |
+
"grad_norm": 0.80078125,
|
| 781 |
+
"learning_rate": 1.948722813130624e-05,
|
| 782 |
+
"loss": 0.023038258543238044,
|
| 783 |
+
"num_pairs": 27.5,
|
| 784 |
+
"step": 86
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"avg_label": 0.4539930522441864,
|
| 788 |
+
"epoch": 0.3020833333333333,
|
| 789 |
+
"grad_norm": 1.6328125,
|
| 790 |
+
"learning_rate": 1.9468916617225814e-05,
|
| 791 |
+
"loss": 0.058613574132323265,
|
| 792 |
+
"num_pairs": 28.0,
|
| 793 |
+
"step": 87
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"avg_label": 0.4584091007709503,
|
| 797 |
+
"epoch": 0.3055555555555556,
|
| 798 |
+
"grad_norm": 0.388671875,
|
| 799 |
+
"learning_rate": 1.9450292765601287e-05,
|
| 800 |
+
"loss": 0.016918038250878453,
|
| 801 |
+
"num_pairs": 28.0,
|
| 802 |
+
"step": 88
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"avg_label": 0.4232954829931259,
|
| 806 |
+
"epoch": 0.3090277777777778,
|
| 807 |
+
"grad_norm": 1.3359375,
|
| 808 |
+
"learning_rate": 1.94313571907509e-05,
|
| 809 |
+
"loss": 0.060430300422012806,
|
| 810 |
+
"num_pairs": 28.0,
|
| 811 |
+
"step": 89
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"avg_label": 0.5833333134651184,
|
| 815 |
+
"epoch": 0.3125,
|
| 816 |
+
"grad_norm": 4.125,
|
| 817 |
+
"learning_rate": 1.941211051727524e-05,
|
| 818 |
+
"loss": 0.060059962212108076,
|
| 819 |
+
"num_pairs": 20.5,
|
| 820 |
+
"step": 90
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"avg_label": 0.45364584028720856,
|
| 824 |
+
"epoch": 0.3159722222222222,
|
| 825 |
+
"grad_norm": 1.4375,
|
| 826 |
+
"learning_rate": 1.939255338003666e-05,
|
| 827 |
+
"loss": 0.04583445412572473,
|
| 828 |
+
"num_pairs": 28.0,
|
| 829 |
+
"step": 91
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"avg_label": 0.4437500089406967,
|
| 833 |
+
"epoch": 0.3194444444444444,
|
| 834 |
+
"grad_norm": 0.26171875,
|
| 835 |
+
"learning_rate": 1.937268642413835e-05,
|
| 836 |
+
"loss": 0.00882079591974616,
|
| 837 |
+
"num_pairs": 27.0,
|
| 838 |
+
"step": 92
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"avg_label": 0.4508301317691803,
|
| 842 |
+
"epoch": 0.3229166666666667,
|
| 843 |
+
"grad_norm": 0.51953125,
|
| 844 |
+
"learning_rate": 1.9352510304903017e-05,
|
| 845 |
+
"loss": 0.03392870319657959,
|
| 846 |
+
"num_pairs": 27.5,
|
| 847 |
+
"step": 93
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"avg_label": 0.44549499452114105,
|
| 851 |
+
"epoch": 0.3263888888888889,
|
| 852 |
+
"grad_norm": 1.4375,
|
| 853 |
+
"learning_rate": 1.9332025687851325e-05,
|
| 854 |
+
"loss": 0.04178585158661008,
|
| 855 |
+
"num_pairs": 28.0,
|
| 856 |
+
"step": 94
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"avg_label": 0.47159090638160706,
|
| 860 |
+
"epoch": 0.3298611111111111,
|
| 861 |
+
"grad_norm": 0.92578125,
|
| 862 |
+
"learning_rate": 1.931123324867989e-05,
|
| 863 |
+
"loss": 0.028863655403256416,
|
| 864 |
+
"num_pairs": 28.0,
|
| 865 |
+
"step": 95
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"avg_label": 0.4557291716337204,
|
| 869 |
+
"epoch": 0.3333333333333333,
|
| 870 |
+
"grad_norm": 2.671875,
|
| 871 |
+
"learning_rate": 1.929013367323902e-05,
|
| 872 |
+
"loss": 0.1290583610534668,
|
| 873 |
+
"num_pairs": 28.0,
|
| 874 |
+
"step": 96
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"avg_label": 0.44440363347530365,
|
| 878 |
+
"epoch": 0.3368055555555556,
|
| 879 |
+
"grad_norm": 0.75390625,
|
| 880 |
+
"learning_rate": 1.926872765751009e-05,
|
| 881 |
+
"loss": 0.01571450149640441,
|
| 882 |
+
"num_pairs": 28.0,
|
| 883 |
+
"step": 97
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"avg_label": 0.5104166716337204,
|
| 887 |
+
"epoch": 0.3402777777777778,
|
| 888 |
+
"grad_norm": 5.9375,
|
| 889 |
+
"learning_rate": 1.9247015907582574e-05,
|
| 890 |
+
"loss": 0.2692837491631508,
|
| 891 |
+
"num_pairs": 27.5,
|
| 892 |
+
"step": 98
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"avg_label": 0.4975000023841858,
|
| 896 |
+
"epoch": 0.34375,
|
| 897 |
+
"grad_norm": 0.462890625,
|
| 898 |
+
"learning_rate": 1.9224999139630766e-05,
|
| 899 |
+
"loss": 0.02842018473893404,
|
| 900 |
+
"num_pairs": 24.0,
|
| 901 |
+
"step": 99
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"avg_label": 0.46345899999141693,
|
| 905 |
+
"epoch": 0.3472222222222222,
|
| 906 |
+
"grad_norm": 3.578125,
|
| 907 |
+
"learning_rate": 1.920267807989015e-05,
|
| 908 |
+
"loss": 0.12233215570449829,
|
| 909 |
+
"num_pairs": 25.5,
|
| 910 |
+
"step": 100
|
| 911 |
+
}
|
| 912 |
+
],
|
| 913 |
+
"logging_steps": 1.0,
|
| 914 |
+
"max_steps": 576,
|
| 915 |
+
"num_input_tokens_seen": 0,
|
| 916 |
+
"num_train_epochs": 2,
|
| 917 |
+
"save_steps": 100,
|
| 918 |
+
"stateful_callbacks": {
|
| 919 |
+
"TrainerControl": {
|
| 920 |
+
"args": {
|
| 921 |
+
"should_epoch_stop": false,
|
| 922 |
+
"should_evaluate": false,
|
| 923 |
+
"should_log": false,
|
| 924 |
+
"should_save": true,
|
| 925 |
+
"should_training_stop": false
|
| 926 |
+
},
|
| 927 |
+
"attributes": {}
|
| 928 |
+
}
|
| 929 |
+
},
|
| 930 |
+
"total_flos": 0.0,
|
| 931 |
+
"train_batch_size": 32,
|
| 932 |
+
"trial_name": null,
|
| 933 |
+
"trial_params": null
|
| 934 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec0a2c7e7e550f7fac2bfb9f535d02f16d9428695d17f831a7aaf9a829592c3f
|
| 3 |
+
size 5496
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|