Hugging Face's logo

lihy285
/
GeoThinker

Safetensors
Model card Files
xet
 Community
1
lihy285 commited on
Commit
484f036
·
verified ·
1 Parent(s): 909ce3a

Delete QeoThinker-VGGT-Qwen25VL-7B-Vanilla

Browse files
Files changed (18) hide show
  1. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/added_tokens.json +0 -24
  2. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/chat_template.json +0 -3
  3. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/config.json +0 -87
  4. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/generation_config.json +0 -12
  5. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/merges.txt +0 -0
  6. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00001-of-00005.safetensors +0 -3
  7. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00002-of-00005.safetensors +0 -3
  8. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00003-of-00005.safetensors +0 -3
  9. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00004-of-00005.safetensors +0 -3
  10. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00005-of-00005.safetensors +0 -3
  11. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model.safetensors.index.json +0 -0
  12. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/preprocessor_config.json +0 -29
  13. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/special_tokens_map.json +0 -31
  14. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/tokenizer_config.json +0 -209
  15. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/train.log +0 -928
  16. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/trainer_state.json +0 -3298
  17. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/training_args.bin +0 -3
  18. QeoThinker-VGGT-Qwen25VL-7B-Vanilla/vocab.json +0 -0
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/added_tokens.json DELETED
@@ -1,24 +0,0 @@
1
- {
2
- "</tool_call>": 151658,
3
- "<tool_call>": 151657,
4
- "<|box_end|>": 151649,
5
- "<|box_start|>": 151648,
6
- "<|endoftext|>": 151643,
7
- "<|file_sep|>": 151664,
8
- "<|fim_middle|>": 151660,
9
- "<|fim_pad|>": 151662,
10
- "<|fim_prefix|>": 151659,
11
- "<|fim_suffix|>": 151661,
12
- "<|im_end|>": 151645,
13
- "<|im_start|>": 151644,
14
- "<|image_pad|>": 151655,
15
- "<|object_ref_end|>": 151647,
16
- "<|object_ref_start|>": 151646,
17
- "<|quad_end|>": 151651,
18
- "<|quad_start|>": 151650,
19
- "<|repo_name|>": 151663,
20
- "<|video_pad|>": 151656,
21
- "<|vision_end|>": 151653,
22
- "<|vision_pad|>": 151654,
23
- "<|vision_start|>": 151652
24
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/chat_template.json DELETED
@@ -1,3 +0,0 @@
1
- {
2
- "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
- }
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/config.json DELETED
@@ -1,87 +0,0 @@
1
- {
2
- "align_method": "zero",
3
- "align_method_weight": 0.1,
4
- "architectures": [
5
- "Qwen2_5_VLForConditionalGenerationWithVGGT"
6
- ],
7
- "attention_dropout": 0.0,
8
- "bos_token_id": 151643,
9
- "cam_merger_type": "zero",
10
- "depart_smi_token": false,
11
- "eos_token_id": 151645,
12
- "feature_fusion_method": "zero",
13
- "fusion_num_layers": 1,
14
- "geo_cross_attn": true,
15
- "geo_importance_gate": true,
16
- "geo_inject_version": "v1",
17
- "geo_layer_interval": 1,
18
- "geo_learn_bias": false,
19
- "geo_spatial_bias": false,
20
- "geometry_encoder_type": "vggt",
21
- "geometry_merger_type": "mlp",
22
- "hidden_act": "silu",
23
- "hidden_size": 3584,
24
- "image_token_id": 151655,
25
- "initializer_range": 0.02,
26
- "intermediate_size": 18944,
27
- "max_position_embeddings": 128000,
28
- "max_window_layers": 28,
29
- "model_type": "qwen2_5_vl",
30
- "num_attention_heads": 28,
31
- "num_hidden_layers": 28,
32
- "num_key_value_heads": 4,
33
- "reference_frame": "first",
34
- "rms_norm_eps": 1e-06,
35
- "rope_scaling": {
36
- "mrope_section": [
37
- 16,
38
- 24,
39
- 24
40
- ],
41
- "rope_type": "default",
42
- "type": "default"
43
- },
44
- "rope_theta": 1000000.0,
45
- "selection_method": "zero",
46
- "selection_method_ratio": 0.25,
47
- "sliding_window": 32768,
48
- "smi_downsample_rate": 2,
49
- "smi_image_num": 8,
50
- "tie_word_embeddings": false,
51
- "torch_dtype": "bfloat16",
52
- "training": true,
53
- "transformers_version": "4.50.0",
54
- "use_cache": true,
55
- "use_geometry_encoder": true,
56
- "use_qwenvl_loss": false,
57
- "use_sliding_window": false,
58
- "video_token_id": 151656,
59
- "vision_config": {
60
- "depth": 32,
61
- "fullatt_block_indexes": [
62
- 7,
63
- 15,
64
- 23,
65
- 31
66
- ],
67
- "hidden_act": "silu",
68
- "hidden_size": 1280,
69
- "in_channels": 3,
70
- "in_chans": 3,
71
- "intermediate_size": 3420,
72
- "model_type": "qwen2_5_vl",
73
- "num_heads": 16,
74
- "out_hidden_size": 3584,
75
- "patch_size": 14,
76
- "spatial_merge_size": 2,
77
- "spatial_patch_size": 14,
78
- "temporal_patch_size": 2,
79
- "tokens_per_second": 2,
80
- "torch_dtype": "bfloat16",
81
- "window_size": 112
82
- },
83
- "vision_end_token_id": 151653,
84
- "vision_start_token_id": 151652,
85
- "vision_token_id": 151654,
86
- "vocab_size": 152064
87
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/generation_config.json DELETED
@@ -1,12 +0,0 @@
1
- {
2
- "bos_token_id": 151643,
3
- "do_sample": true,
4
- "eos_token_id": [
5
- 151645,
6
- 151643
7
- ],
8
- "pad_token_id": 151643,
9
- "repetition_penalty": 1.05,
10
- "temperature": 1e-06,
11
- "transformers_version": "4.50.0"
12
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/merges.txt DELETED
The diff for this file is too large to render. See raw diff
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00001-of-00005.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:70ce11bb19d95f66f54c8c997ce0f8a3587c7680976b9e9fcc6bbcc359d03a2c
3
- size 4979504134
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00002-of-00005.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:3d2dd11e7ed7cb5f8d383cf84975703fdb7563ad84e8542558db5ebca92a95b3
3
- size 4932982360
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00003-of-00005.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a83756d32444e4617234c69385c4d9c5867721afb39dc23dd648197b39720334
3
- size 4906423326
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00004-of-00005.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:f81aba14b898a4149058edbe692ad9673a87c386a234ba23b8122ffce6dda477
3
- size 4874252312
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model-00005-of-00005.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a3ef632569857d037ce42eda45f2192991cc44368c507b15214a3d73db772c9a
3
- size 1910358204
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/model.safetensors.index.json DELETED
The diff for this file is too large to render. See raw diff
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/preprocessor_config.json DELETED
@@ -1,29 +0,0 @@
1
- {
2
- "do_convert_rgb": true,
3
- "do_normalize": true,
4
- "do_rescale": true,
5
- "do_resize": true,
6
- "image_mean": [
7
- 0.48145466,
8
- 0.4578275,
9
- 0.40821073
10
- ],
11
- "image_processor_type": "Qwen2VLImageProcessor",
12
- "image_std": [
13
- 0.26862954,
14
- 0.26130258,
15
- 0.27577711
16
- ],
17
- "max_pixels": 451584,
18
- "merge_size": 2,
19
- "min_pixels": 12544,
20
- "patch_size": 14,
21
- "processor_class": "Qwen2_5_VLProcessor",
22
- "resample": 3,
23
- "rescale_factor": 0.00392156862745098,
24
- "size": {
25
- "longest_edge": 451584,
26
- "shortest_edge": 12544
27
- },
28
- "temporal_patch_size": 2
29
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/special_tokens_map.json DELETED
@@ -1,31 +0,0 @@
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
- ],
17
- "eos_token": {
18
- "content": "<|im_end|>",
19
- "lstrip": false,
20
- "normalized": false,
21
- "rstrip": false,
22
- "single_word": false
23
- },
24
- "pad_token": {
25
- "content": "<|endoftext|>",
26
- "lstrip": false,
27
- "normalized": false,
28
- "rstrip": false,
29
- "single_word": false
30
- }
31
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/tokenizer_config.json DELETED
@@ -1,209 +0,0 @@
1
- {
2
- "add_bos_token": false,
3
- "add_prefix_space": false,
4
- "added_tokens_decoder": {
5
- "151643": {
6
- "content": "<|endoftext|>",
7
- "lstrip": false,
8
- "normalized": false,
9
- "rstrip": false,
10
- "single_word": false,
11
- "special": true
12
- },
13
- "151644": {
14
- "content": "<|im_start|>",
15
- "lstrip": false,
16
- "normalized": false,
17
- "rstrip": false,
18
- "single_word": false,
19
- "special": true
20
- },
21
- "151645": {
22
- "content": "<|im_end|>",
23
- "lstrip": false,
24
- "normalized": false,
25
- "rstrip": false,
26
- "single_word": false,
27
- "special": true
28
- },
29
- "151646": {
30
- "content": "<|object_ref_start|>",
31
- "lstrip": false,
32
- "normalized": false,
33
- "rstrip": false,
34
- "single_word": false,
35
- "special": true
36
- },
37
- "151647": {
38
- "content": "<|object_ref_end|>",
39
- "lstrip": false,
40
- "normalized": false,
41
- "rstrip": false,
42
- "single_word": false,
43
- "special": true
44
- },
45
- "151648": {
46
- "content": "<|box_start|>",
47
- "lstrip": false,
48
- "normalized": false,
49
- "rstrip": false,
50
- "single_word": false,
51
- "special": true
52
- },
53
- "151649": {
54
- "content": "<|box_end|>",
55
- "lstrip": false,
56
- "normalized": false,
57
- "rstrip": false,
58
- "single_word": false,
59
- "special": true
60
- },
61
- "151650": {
62
- "content": "<|quad_start|>",
63
- "lstrip": false,
64
- "normalized": false,
65
- "rstrip": false,
66
- "single_word": false,
67
- "special": true
68
- },
69
- "151651": {
70
- "content": "<|quad_end|>",
71
- "lstrip": false,
72
- "normalized": false,
73
- "rstrip": false,
74
- "single_word": false,
75
- "special": true
76
- },
77
- "151652": {
78
- "content": "<|vision_start|>",
79
- "lstrip": false,
80
- "normalized": false,
81
- "rstrip": false,
82
- "single_word": false,
83
- "special": true
84
- },
85
- "151653": {
86
- "content": "<|vision_end|>",
87
- "lstrip": false,
88
- "normalized": false,
89
- "rstrip": false,
90
- "single_word": false,
91
- "special": true
92
- },
93
- "151654": {
94
- "content": "<|vision_pad|>",
95
- "lstrip": false,
96
- "normalized": false,
97
- "rstrip": false,
98
- "single_word": false,
99
- "special": true
100
- },
101
- "151655": {
102
- "content": "<|image_pad|>",
103
- "lstrip": false,
104
- "normalized": false,
105
- "rstrip": false,
106
- "single_word": false,
107
- "special": true
108
- },
109
- "151656": {
110
- "content": "<|video_pad|>",
111
- "lstrip": false,
112
- "normalized": false,
113
- "rstrip": false,
114
- "single_word": false,
115
- "special": true
116
- },
117
- "151657": {
118
- "content": "<tool_call>",
119
- "lstrip": false,
120
- "normalized": false,
121
- "rstrip": false,
122
- "single_word": false,
123
- "special": false
124
- },
125
- "151658": {
126
- "content": "</tool_call>",
127
- "lstrip": false,
128
- "normalized": false,
129
- "rstrip": false,
130
- "single_word": false,
131
- "special": false
132
- },
133
- "151659": {
134
- "content": "<|fim_prefix|>",
135
- "lstrip": false,
136
- "normalized": false,
137
- "rstrip": false,
138
- "single_word": false,
139
- "special": false
140
- },
141
- "151660": {
142
- "content": "<|fim_middle|>",
143
- "lstrip": false,
144
- "normalized": false,
145
- "rstrip": false,
146
- "single_word": false,
147
- "special": false
148
- },
149
- "151661": {
150
- "content": "<|fim_suffix|>",
151
- "lstrip": false,
152
- "normalized": false,
153
- "rstrip": false,
154
- "single_word": false,
155
- "special": false
156
- },
157
- "151662": {
158
- "content": "<|fim_pad|>",
159
- "lstrip": false,
160
- "normalized": false,
161
- "rstrip": false,
162
- "single_word": false,
163
- "special": false
164
- },
165
- "151663": {
166
- "content": "<|repo_name|>",
167
- "lstrip": false,
168
- "normalized": false,
169
- "rstrip": false,
170
- "single_word": false,
171
- "special": false
172
- },
173
- "151664": {
174
- "content": "<|file_sep|>",
175
- "lstrip": false,
176
- "normalized": false,
177
- "rstrip": false,
178
- "single_word": false,
179
- "special": false
180
- }
181
- },
182
- "additional_special_tokens": [
183
- "<|im_start|>",
184
- "<|im_end|>",
185
- "<|object_ref_start|>",
186
- "<|object_ref_end|>",
187
- "<|box_start|>",
188
- "<|box_end|>",
189
- "<|quad_start|>",
190
- "<|quad_end|>",
191
- "<|vision_start|>",
192
- "<|vision_end|>",
193
- "<|vision_pad|>",
194
- "<|image_pad|>",
195
- "<|video_pad|>"
196
- ],
197
- "bos_token": null,
198
- "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
199
- "clean_up_tokenization_spaces": false,
200
- "eos_token": "<|im_end|>",
201
- "errors": "replace",
202
- "extra_special_tokens": {},
203
- "model_max_length": 12800,
204
- "pad_token": "<|endoftext|>",
205
- "padding_side": "right",
206
- "split_special_tokens": false,
207
- "tokenizer_class": "Qwen2Tokenizer",
208
- "unk_token": null
209
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/train.log DELETED
@@ -1,928 +0,0 @@
1
- Trainer._get_train_sampler replaced with custom implementation.
2
- load moxing failed
3
- Trainer._get_train_sampler replaced with custom implementation.
4
- load moxing failed
5
- [2026-01-14 23:17:48,691] [INFO] [real_accelerator.py:222:get_accelerator] Setting ds_accelerator to cuda (auto detect)
6
- Trainer._get_train_sampler replaced with custom implementation.
7
- load moxing failed
8
- Trainer._get_train_sampler replaced with custom implementation.
9
- Trainer._get_train_sampler replaced with custom implementation.
10
- load moxing failed
11
- load moxing failed
12
- Trainer._get_train_sampler replaced with custom implementation.
13
- Trainer._get_train_sampler replaced with custom implementation.
14
- load moxing failed
15
- load moxing failed
16
- Trainer._get_train_sampler replaced with custom implementation.
17
- load moxing failed
18
- [2026-01-14 23:17:49,114] [INFO] [real_accelerator.py:222:get_accelerator] Setting ds_accelerator to cuda (auto detect)
19
- [2026-01-14 23:17:49,227] [INFO] [real_accelerator.py:222:get_accelerator] Setting ds_accelerator to cuda (auto detect)
20
- [2026-01-14 23:17:49,250] [INFO] [real_accelerator.py:222:get_accelerator] Setting ds_accelerator to cuda (auto detect)
21
- [2026-01-14 23:17:49,269] [INFO] [real_accelerator.py:222:get_accelerator] Setting ds_accelerator to cuda (auto detect)
22
- [2026-01-14 23:17:49,275] [INFO] [real_accelerator.py:222:get_accelerator] Setting ds_accelerator to cuda (auto detect)
23
- [2026-01-14 23:17:49,275] [INFO] [real_accelerator.py:222:get_accelerator] Setting ds_accelerator to cuda (auto detect)
24
- [2026-01-14 23:17:49,279] [INFO] [real_accelerator.py:222:get_accelerator] Setting ds_accelerator to cuda (auto detect)
25
- [2026-01-14 23:17:49,620] [INFO] [comm.py:658:init_distributed] cdb=None
26
- Warning: FlashAttention 3 is not available, falling back to PyTorch's scaled_dot_product_attention
27
- /cache/pretrained/Qwen2.5-VL-7B-Instruct/
28
- [2026-01-14 23:17:49,980] [INFO] [comm.py:658:init_distributed] cdb=None
29
- [2026-01-14 23:17:50,125] [INFO] [comm.py:658:init_distributed] cdb=None
30
- [2026-01-14 23:17:50,163] [INFO] [comm.py:658:init_distributed] cdb=None
31
- [2026-01-14 23:17:50,350] [INFO] [comm.py:658:init_distributed] cdb=None
32
- [2026-01-14 23:17:50,357] [INFO] [comm.py:658:init_distributed] cdb=None
33
- [2026-01-14 23:17:50,431] [INFO] [comm.py:658:init_distributed] cdb=None
34
- [2026-01-14 23:17:50,469] [INFO] [comm.py:658:init_distributed] cdb=None
35
- Warning: FlashAttention 3 is not available, falling back to PyTorch's scaled_dot_product_attention
36
- Warning: FlashAttention 3 is not available, falling back to PyTorch's scaled_dot_product_attention
37
- Warning: FlashAttention 3 is not available, falling back to PyTorch's scaled_dot_product_attention
38
- Warning: FlashAttention 3 is not available, falling back to PyTorch's scaled_dot_product_attention
39
- /cache/pretrained/Qwen2.5-VL-7B-Instruct/
40
- /cache/pretrained/Qwen2.5-VL-7B-Instruct/
41
- /cache/pretrained/Qwen2.5-VL-7B-Instruct/
42
- /cache/pretrained/Qwen2.5-VL-7B-Instruct/
43
- Warning: FlashAttention 3 is not available, falling back to PyTorch's scaled_dot_product_attention
44
- Warning: FlashAttention 3 is not available, falling back to PyTorch's scaled_dot_product_attention
45
- Warning: FlashAttention 3 is not available, falling back to PyTorch's scaled_dot_product_attention
46
- /cache/pretrained/Qwen2.5-VL-7B-Instruct/
47
- /cache/pretrained/Qwen2.5-VL-7B-Instruct/
48
- /cache/pretrained/Qwen2.5-VL-7B-Instruct/
49
- fusion_config FeatureFusionConfig(fusion_method='zero', hidden_size=3584, num_heads=8, dropout=0.1, num_layers=1, cam_merger_type='zero', align_method='zero', selection_method='zero', selection_method_ratio=0.25, align_method_weight=0.1, training=True)
50
- fusion_config FeatureFusionConfig(fusion_method='zero', hidden_size=3584, num_heads=8, dropout=0.1, num_layers=1, cam_merger_type='zero', align_method='zero', selection_method='zero', selection_method_ratio=0.25, align_method_weight=0.1, training=True)
51
- fusion_config FeatureFusionConfig(fusion_method='zero', hidden_size=3584, num_heads=8, dropout=0.1, num_layers=1, cam_merger_type='zero', align_method='zero', selection_method='zero', selection_method_ratio=0.25, align_method_weight=0.1, training=True)
52
- fusion_config FeatureFusionConfig(fusion_method='zero', hidden_size=3584, num_heads=8, dropout=0.1, num_layers=1, cam_merger_type='zero', align_method='zero', selection_method='zero', selection_method_ratio=0.25, align_method_weight=0.1, training=True)
53
- fusion_config FeatureFusionConfig(fusion_method='zero', hidden_size=3584, num_heads=8, dropout=0.1, num_layers=1, cam_merger_type='zero', align_method='zero', selection_method='zero', selection_method_ratio=0.25, align_method_weight=0.1, training=True)
54
- fusion_config FeatureFusionConfig(fusion_method='zero', hidden_size=3584, num_heads=8, dropout=0.1, num_layers=1, cam_merger_type='zero', align_method='zero', selection_method='zero', selection_method_ratio=0.25, align_method_weight=0.1, training=True)
55
- fusion_config FeatureFusionConfig(fusion_method='zero', hidden_size=3584, num_heads=8, dropout=0.1, num_layers=1, cam_merger_type='zero', align_method='zero', selection_method='zero', selection_method_ratio=0.25, align_method_weight=0.1, training=True)
56
- fusion_config FeatureFusionConfig(fusion_method='zero', hidden_size=3584, num_heads=8, dropout=0.1, num_layers=1, cam_merger_type='zero', align_method='zero', selection_method='zero', selection_method_ratio=0.25, align_method_weight=0.1, training=True)
57
- Loading weights from local directory
58
- Loading weights from local directory
59
- Loading weights from local directory
60
- Loading weights from local directory
61
- Loading weights from local directory
62
- Loading weights from local directory
63
- Loading weights from local directory
64
- Loading weights from local directory
65
- dataset_names ['llava_hound_64k', 'spar_234k']
66
- Loading datasets: [{'annotation_path': 'data/train/llava_hound_64k.json', 'data_path': 'data/media', 'tag': '2d', 'sampling_rate': 1.0, 'dataset_name': 'llava_hound_64k'}, {'annotation_path': 'data/train/spar_234k.json', 'data_path': 'data/media', 'tag': '3d', 'sampling_rate': 1.0, 'dataset_name': 'spar_234k'}]
67
- dataset_names ['llava_hound_64k', 'spar_234k']
68
- Loading datasets: [{'annotation_path': 'data/train/llava_hound_64k.json', 'data_path': 'data/media', 'tag': '2d', 'sampling_rate': 1.0, 'dataset_name': 'llava_hound_64k'}, {'annotation_path': 'data/train/spar_234k.json', 'data_path': 'data/media', 'tag': '3d', 'sampling_rate': 1.0, 'dataset_name': 'spar_234k'}]
69
- dataset_names ['llava_hound_64k', 'spar_234k']
70
- Loading datasets: [{'annotation_path': 'data/train/llava_hound_64k.json', 'data_path': 'data/media', 'tag': '2d', 'sampling_rate': 1.0, 'dataset_name': 'llava_hound_64k'}, {'annotation_path': 'data/train/spar_234k.json', 'data_path': 'data/media', 'tag': '3d', 'sampling_rate': 1.0, 'dataset_name': 'spar_234k'}]
71
- Qwen2_5_VLConfig {
72
- "_attn_implementation_autoset": true,
73
- "align_method": "zero",
74
- "align_method_weight": 0.1,
75
- "architectures": [
76
- "Qwen2_5_VLForConditionalGeneration"
77
- ],
78
- "attention_dropout": 0.0,
79
- "bos_token_id": 151643,
80
- "cam_merger_type": "zero",
81
- "depart_smi_token": false,
82
- "eos_token_id": 151645,
83
- "feature_fusion_method": "zero",
84
- "fusion_num_layers": 1,
85
- "geo_cross_attn": true,
86
- "geo_importance_gate": true,
87
- "geo_inject_version": "v4",
88
- "geo_layer_interval": 1,
89
- "geo_learn_bias": false,
90
- "geo_spatial_bias": false,
91
- "geometry_encoder_type": "vggt",
92
- "geometry_merger_type": "mlp",
93
- "hidden_act": "silu",
94
- "hidden_size": 3584,
95
- "image_token_id": 151655,
96
- "initializer_range": 0.02,
97
- "intermediate_size": 18944,
98
- "max_position_embeddings": 128000,
99
- "max_window_layers": 28,
100
- "model_type": "qwen2_5_vl",
101
- "num_attention_heads": 28,
102
- "num_hidden_layers": 28,
103
- "num_key_value_heads": 4,
104
- "reference_frame": "first",
105
- "rms_norm_eps": 1e-06,
106
- "rope_scaling": {
107
- "mrope_section": [
108
- 16,
109
- 24,
110
- 24
111
- ],
112
- "rope_type": "default",
113
- "type": "default"
114
- },
115
- "rope_theta": 1000000.0,
116
- "selection_method": "zero",
117
- "selection_method_ratio": 0.25,
118
- "sliding_window": 32768,
119
- "smi_downsample_rate": 2,
120
- "smi_image_num": 8,
121
- "tie_word_embeddings": false,
122
- "torch_dtype": "bfloat16",
123
- "training": true,
124
- "transformers_version": "4.50.0",
125
- "use_cache": false,
126
- "use_geometry_encoder": true,
127
- "use_qwenvl_loss": false,
128
- "use_sliding_window": false,
129
- "video_token_id": 151656,
130
- "vision_config": {
131
- "depth": 32,
132
- "fullatt_block_indexes": [
133
- 7,
134
- 15,
135
- 23,
136
- 31
137
- ],
138
- "hidden_act": "silu",
139
- "hidden_size": 1280,
140
- "in_channels": 3,
141
- "in_chans": 3,
142
- "intermediate_size": 3420,
143
- "model_type": "qwen2_5_vl",
144
- "num_heads": 16,
145
- "out_hidden_size": 3584,
146
- "patch_size": 14,
147
- "spatial_merge_size": 2,
148
- "spatial_patch_size": 14,
149
- "temporal_patch_size": 2,
150
- "tokens_per_second": 2,
151
- "torch_dtype": "bfloat16",
152
- "window_size": 112
153
- },
154
- "vision_end_token_id": 151653,
155
- "vision_start_token_id": 151652,
156
- "vision_token_id": 151654,
157
- "vocab_size": 152064
158
- }
159
-
160
- ==== ====
161
- ==== Only training the following parameters ====
162
- ==== ====
163
- geometry_merger.ln_q.weight torch.Size([2048])
164
- geometry_merger.mlp.0.weight torch.Size([4096, 8192])
165
- geometry_merger.mlp.0.bias torch.Size([4096])
166
- geometry_merger.mlp.2.weight torch.Size([3584, 4096])
167
- geometry_merger.mlp.2.bias torch.Size([3584])
168
- geometry_merger.camera_mlp.0.weight torch.Size([4096, 2048])
169
- geometry_merger.camera_mlp.0.bias torch.Size([4096])
170
- geometry_merger.camera_mlp.2.weight torch.Size([3584, 4096])
171
- geometry_merger.camera_mlp.2.bias torch.Size([3584])
172
- model.embed_tokens.weight torch.Size([152064, 3584])
173
- model.layers.0.self_attn.q_proj.weight torch.Size([3584, 3584])
174
- model.layers.0.self_attn.q_proj.bias torch.Size([3584])
175
- model.layers.0.self_attn.k_proj.weight torch.Size([512, 3584])
176
- model.layers.0.self_attn.k_proj.bias torch.Size([512])
177
- model.layers.0.self_attn.v_proj.weight torch.Size([512, 3584])
178
- model.layers.0.self_attn.v_proj.bias torch.Size([512])
179
- model.layers.0.self_attn.o_proj.weight torch.Size([3584, 3584])
180
- model.layers.0.mlp.gate_proj.weight torch.Size([18944, 3584])
181
- model.layers.0.mlp.up_proj.weight torch.Size([18944, 3584])
182
- model.layers.0.mlp.down_proj.weight torch.Size([3584, 18944])
183
- model.layers.0.input_layernorm.weight torch.Size([3584])
184
- model.layers.0.post_attention_layernorm.weight torch.Size([3584])
185
- model.layers.0.cross_attn.gate torch.Size([])
186
- model.layers.0.cross_attn.spatial_scale_param torch.Size([])
187
- model.layers.0.cross_attn.input_layernorm.weight torch.Size([3584])
188
- model.layers.0.cross_attn.input_layernorm.bias torch.Size([3584])
189
- model.layers.0.cross_attn.vggt_layernorm.weight torch.Size([3584])
190
- model.layers.0.cross_attn.vggt_layernorm.bias torch.Size([3584])
191
- model.layers.0.cross_attn.q_proj.weight torch.Size([3584, 3584])
192
- model.layers.0.cross_attn.k_proj.weight torch.Size([3584, 3584])
193
- model.layers.0.cross_attn.v_proj.weight torch.Size([3584, 3584])
194
- model.layers.0.cross_attn.o_proj.weight torch.Size([3584, 3584])
195
- model.layers.0.cross_attn.importance_net.0.weight torch.Size([896, 3584])
196
- model.layers.0.cross_attn.importance_net.0.bias torch.Size([896])
197
- model.layers.0.cross_attn.importance_net.2.weight torch.Size([1, 896])
198
- model.layers.0.cross_attn.importance_net.2.bias torch.Size([1])
199
- model.layers.1.self_attn.q_proj.weight torch.Size([3584, 3584])
200
- model.layers.1.self_attn.q_proj.bias torch.Size([3584])
201
- model.layers.1.self_attn.k_proj.weight torch.Size([512, 3584])
202
- model.layers.1.self_attn.k_proj.bias torch.Size([512])
203
- model.layers.1.self_attn.v_proj.weight torch.Size([512, 3584])
204
- model.layers.1.self_attn.v_proj.bias torch.Size([512])
205
- model.layers.1.self_attn.o_proj.weight torch.Size([3584, 3584])
206
- model.layers.1.mlp.gate_proj.weight torch.Size([18944, 3584])
207
- model.layers.1.mlp.up_proj.weight torch.Size([18944, 3584])
208
- model.layers.1.mlp.down_proj.weight torch.Size([3584, 18944])
209
- model.layers.1.input_layernorm.weight torch.Size([3584])
210
- model.layers.1.post_attention_layernorm.weight torch.Size([3584])
211
- model.layers.1.cross_attn.gate torch.Size([])
212
- model.layers.1.cross_attn.spatial_scale_param torch.Size([])
213
- model.layers.1.cross_attn.input_layernorm.weight torch.Size([3584])
214
- model.layers.1.cross_attn.input_layernorm.bias torch.Size([3584])
215
- model.layers.1.cross_attn.vggt_layernorm.weight torch.Size([3584])
216
- model.layers.1.cross_attn.vggt_layernorm.bias torch.Size([3584])
217
- model.layers.1.cross_attn.q_proj.weight torch.Size([3584, 3584])
218
- model.layers.1.cross_attn.k_proj.weight torch.Size([3584, 3584])
219
- model.layers.1.cross_attn.v_proj.weight torch.Size([3584, 3584])
220
- model.layers.1.cross_attn.o_proj.weight torch.Size([3584, 3584])
221
- model.layers.1.cross_attn.importance_net.0.weight torch.Size([896, 3584])
222
- model.layers.1.cross_attn.importance_net.0.bias torch.Size([896])
223
- model.layers.1.cross_attn.importance_net.2.weight torch.Size([1, 896])
224
- model.layers.1.cross_attn.importance_net.2.bias torch.Size([1])
225
- model.layers.2.self_attn.q_proj.weight torch.Size([3584, 3584])
226
- model.layers.2.self_attn.q_proj.bias torch.Size([3584])
227
- model.layers.2.self_attn.k_proj.weight torch.Size([512, 3584])
228
- model.layers.2.self_attn.k_proj.bias torch.Size([512])
229
- model.layers.2.self_attn.v_proj.weight torch.Size([512, 3584])
230
- model.layers.2.self_attn.v_proj.bias torch.Size([512])
231
- model.layers.2.self_attn.o_proj.weight torch.Size([3584, 3584])
232
- model.layers.2.mlp.gate_proj.weight torch.Size([18944, 3584])
233
- model.layers.2.mlp.up_proj.weight torch.Size([18944, 3584])
234
- model.layers.2.mlp.down_proj.weight torch.Size([3584, 18944])
235
- model.layers.2.input_layernorm.weight torch.Size([3584])
236
- model.layers.2.post_attention_layernorm.weight torch.Size([3584])
237
- model.layers.2.cross_attn.gate torch.Size([])
238
- model.layers.2.cross_attn.spatial_scale_param torch.Size([])
239
- model.layers.2.cross_attn.input_layernorm.weight torch.Size([3584])
240
- model.layers.2.cross_attn.input_layernorm.bias torch.Size([3584])
241
- model.layers.2.cross_attn.vggt_layernorm.weight torch.Size([3584])
242
- model.layers.2.cross_attn.vggt_layernorm.bias torch.Size([3584])
243
- model.layers.2.cross_attn.q_proj.weight torch.Size([3584, 3584])
244
- model.layers.2.cross_attn.k_proj.weight torch.Size([3584, 3584])
245
- model.layers.2.cross_attn.v_proj.weight torch.Size([3584, 3584])
246
- model.layers.2.cross_attn.o_proj.weight torch.Size([3584, 3584])
247
- model.layers.2.cross_attn.importance_net.0.weight torch.Size([896, 3584])
248
- model.layers.2.cross_attn.importance_net.0.bias torch.Size([896])
249
- model.layers.2.cross_attn.importance_net.2.weight torch.Size([1, 896])
250
- model.layers.2.cross_attn.importance_net.2.bias torch.Size([1])
251
- model.layers.3.self_attn.q_proj.weight torch.Size([3584, 3584])
252
- model.layers.3.self_attn.q_proj.bias torch.Size([3584])
253
- model.layers.3.self_attn.k_proj.weight torch.Size([512, 3584])
254
- model.layers.3.self_attn.k_proj.bias torch.Size([512])
255
- model.layers.3.self_attn.v_proj.weight torch.Size([512, 3584])
256
- model.layers.3.self_attn.v_proj.bias torch.Size([512])
257
- model.layers.3.self_attn.o_proj.weight torch.Size([3584, 3584])
258
- model.layers.3.mlp.gate_proj.weight torch.Size([18944, 3584])
259
- model.layers.3.mlp.up_proj.weight torch.Size([18944, 3584])
260
- model.layers.3.mlp.down_proj.weight torch.Size([3584, 18944])
261
- model.layers.3.input_layernorm.weight torch.Size([3584])
262
- model.layers.3.post_attention_layernorm.weight torch.Size([3584])
263
- model.layers.3.cross_attn.gate torch.Size([])
264
- model.layers.3.cross_attn.spatial_scale_param torch.Size([])
265
- model.layers.3.cross_attn.input_layernorm.weight torch.Size([3584])
266
- model.layers.3.cross_attn.input_layernorm.bias torch.Size([3584])
267
- model.layers.3.cross_attn.vggt_layernorm.weight torch.Size([3584])
268
- model.layers.3.cross_attn.vggt_layernorm.bias torch.Size([3584])
269
- model.layers.3.cross_attn.q_proj.weight torch.Size([3584, 3584])
270
- model.layers.3.cross_attn.k_proj.weight torch.Size([3584, 3584])
271
- model.layers.3.cross_attn.v_proj.weight torch.Size([3584, 3584])
272
- model.layers.3.cross_attn.o_proj.weight torch.Size([3584, 3584])
273
- model.layers.3.cross_attn.importance_net.0.weight torch.Size([896, 3584])
274
- model.layers.3.cross_attn.importance_net.0.bias torch.Size([896])
275
- model.layers.3.cross_attn.importance_net.2.weight torch.Size([1, 896])
276
- model.layers.3.cross_attn.importance_net.2.bias torch.Size([1])
277
- model.layers.4.self_attn.q_proj.weight torch.Size([3584, 3584])
278
- model.layers.4.self_attn.q_proj.bias torch.Size([3584])
279
- model.layers.4.self_attn.k_proj.weight torch.Size([512, 3584])
280
- model.layers.4.self_attn.k_proj.bias torch.Size([512])
281
- model.layers.4.self_attn.v_proj.weight torch.Size([512, 3584])
282
- model.layers.4.self_attn.v_proj.bias torch.Size([512])
283
- model.layers.4.self_attn.o_proj.weight torch.Size([3584, 3584])
284
- model.layers.4.mlp.gate_proj.weight torch.Size([18944, 3584])
285
- model.layers.4.mlp.up_proj.weight torch.Size([18944, 3584])
286
- model.layers.4.mlp.down_proj.weight torch.Size([3584, 18944])
287
- model.layers.4.input_layernorm.weight torch.Size([3584])
288
- model.layers.4.post_attention_layernorm.weight torch.Size([3584])
289
- model.layers.4.cross_attn.gate torch.Size([])
290
- model.layers.4.cross_attn.spatial_scale_param torch.Size([])
291
- model.layers.4.cross_attn.input_layernorm.weight torch.Size([3584])
292
- model.layers.4.cross_attn.input_layernorm.bias torch.Size([3584])
293
- model.layers.4.cross_attn.vggt_layernorm.weight torch.Size([3584])
294
- model.layers.4.cross_attn.vggt_layernorm.bias torch.Size([3584])
295
- model.layers.4.cross_attn.q_proj.weight torch.Size([3584, 3584])
296
- model.layers.4.cross_attn.k_proj.weight torch.Size([3584, 3584])
297
- model.layers.4.cross_attn.v_proj.weight torch.Size([3584, 3584])
298
- model.layers.4.cross_attn.o_proj.weight torch.Size([3584, 3584])
299
- model.layers.4.cross_attn.importance_net.0.weight torch.Size([896, 3584])
300
- model.layers.4.cross_attn.importance_net.0.bias torch.Size([896])
301
- model.layers.4.cross_attn.importance_net.2.weight torch.Size([1, 896])
302
- model.layers.4.cross_attn.importance_net.2.bias torch.Size([1])
303
- model.layers.5.self_attn.q_proj.weight torch.Size([3584, 3584])
304
- model.layers.5.self_attn.q_proj.bias torch.Size([3584])
305
- model.layers.5.self_attn.k_proj.weight torch.Size([512, 3584])
306
- model.layers.5.self_attn.k_proj.bias torch.Size([512])
307
- model.layers.5.self_attn.v_proj.weight torch.Size([512, 3584])
308
- model.layers.5.self_attn.v_proj.bias torch.Size([512])
309
- model.layers.5.self_attn.o_proj.weight torch.Size([3584, 3584])
310
- model.layers.5.mlp.gate_proj.weight torch.Size([18944, 3584])
311
- model.layers.5.mlp.up_proj.weight torch.Size([18944, 3584])
312
- model.layers.5.mlp.down_proj.weight torch.Size([3584, 18944])
313
- model.layers.5.input_layernorm.weight torch.Size([3584])
314
- model.layers.5.post_attention_layernorm.weight torch.Size([3584])
315
- model.layers.5.cross_attn.gate torch.Size([])
316
- model.layers.5.cross_attn.spatial_scale_param torch.Size([])
317
- model.layers.5.cross_attn.input_layernorm.weight torch.Size([3584])
318
- model.layers.5.cross_attn.input_layernorm.bias torch.Size([3584])
319
- model.layers.5.cross_attn.vggt_layernorm.weight torch.Size([3584])
320
- model.layers.5.cross_attn.vggt_layernorm.bias torch.Size([3584])
321
- model.layers.5.cross_attn.q_proj.weight torch.Size([3584, 3584])
322
- model.layers.5.cross_attn.k_proj.weight torch.Size([3584, 3584])
323
- model.layers.5.cross_attn.v_proj.weight torch.Size([3584, 3584])
324
- model.layers.5.cross_attn.o_proj.weight torch.Size([3584, 3584])
325
- model.layers.5.cross_attn.importance_net.0.weight torch.Size([896, 3584])
326
- model.layers.5.cross_attn.importance_net.0.bias torch.Size([896])
327
- model.layers.5.cross_attn.importance_net.2.weight torch.Size([1, 896])
328
- model.layers.5.cross_attn.importance_net.2.bias torch.Size([1])
329
- model.layers.6.self_attn.q_proj.weight torch.Size([3584, 3584])
330
- model.layers.6.self_attn.q_proj.bias torch.Size([3584])
331
- model.layers.6.self_attn.k_proj.weight torch.Size([512, 3584])
332
- model.layers.6.self_attn.k_proj.bias torch.Size([512])
333
- model.layers.6.self_attn.v_proj.weight torch.Size([512, 3584])
334
- model.layers.6.self_attn.v_proj.bias torch.Size([512])
335
- model.layers.6.self_attn.o_proj.weight torch.Size([3584, 3584])
336
- model.layers.6.mlp.gate_proj.weight torch.Size([18944, 3584])
337
- model.layers.6.mlp.up_proj.weight torch.Size([18944, 3584])
338
- model.layers.6.mlp.down_proj.weight torch.Size([3584, 18944])
339
- model.layers.6.input_layernorm.weight torch.Size([3584])
340
- model.layers.6.post_attention_layernorm.weight torch.Size([3584])
341
- model.layers.6.cross_attn.gate torch.Size([])
342
- model.layers.6.cross_attn.spatial_scale_param torch.Size([])
343
- model.layers.6.cross_attn.input_layernorm.weight torch.Size([3584])
344
- model.layers.6.cross_attn.input_layernorm.bias torch.Size([3584])
345
- model.layers.6.cross_attn.vggt_layernorm.weight torch.Size([3584])
346
- model.layers.6.cross_attn.vggt_layernorm.bias torch.Size([3584])
347
- model.layers.6.cross_attn.q_proj.weight torch.Size([3584, 3584])
348
- model.layers.6.cross_attn.k_proj.weight torch.Size([3584, 3584])
349
- model.layers.6.cross_attn.v_proj.weight torch.Size([3584, 3584])
350
- model.layers.6.cross_attn.o_proj.weight torch.Size([3584, 3584])
351
- model.layers.6.cross_attn.importance_net.0.weight torch.Size([896, 3584])
352
- model.layers.6.cross_attn.importance_net.0.bias torch.Size([896])
353
- model.layers.6.cross_attn.importance_net.2.weight torch.Size([1, 896])
354
- model.layers.6.cross_attn.importance_net.2.bias torch.Size([1])
355
- model.layers.7.self_attn.q_proj.weight torch.Size([3584, 3584])
356
- model.layers.7.self_attn.q_proj.bias torch.Size([3584])
357
- model.layers.7.self_attn.k_proj.weight torch.Size([512, 3584])
358
- model.layers.7.self_attn.k_proj.bias torch.Size([512])
359
- model.layers.7.self_attn.v_proj.weight torch.Size([512, 3584])
360
- model.layers.7.self_attn.v_proj.bias torch.Size([512])
361
- model.layers.7.self_attn.o_proj.weight torch.Size([3584, 3584])
362
- model.layers.7.mlp.gate_proj.weight torch.Size([18944, 3584])
363
- model.layers.7.mlp.up_proj.weight torch.Size([18944, 3584])
364
- model.layers.7.mlp.down_proj.weight torch.Size([3584, 18944])
365
- model.layers.7.input_layernorm.weight torch.Size([3584])
366
- model.layers.7.post_attention_layernorm.weight torch.Size([3584])
367
- model.layers.7.cross_attn.gate torch.Size([])
368
- model.layers.7.cross_attn.spatial_scale_param torch.Size([])
369
- model.layers.7.cross_attn.input_layernorm.weight torch.Size([3584])
370
- model.layers.7.cross_attn.input_layernorm.bias torch.Size([3584])
371
- model.layers.7.cross_attn.vggt_layernorm.weight torch.Size([3584])
372
- model.layers.7.cross_attn.vggt_layernorm.bias torch.Size([3584])
373
- model.layers.7.cross_attn.q_proj.weight torch.Size([3584, 3584])
374
- model.layers.7.cross_attn.k_proj.weight torch.Size([3584, 3584])
375
- model.layers.7.cross_attn.v_proj.weight torch.Size([3584, 3584])
376
- model.layers.7.cross_attn.o_proj.weight torch.Size([3584, 3584])
377
- model.layers.7.cross_attn.importance_net.0.weight torch.Size([896, 3584])
378
- model.layers.7.cross_attn.importance_net.0.bias torch.Size([896])
379
- model.layers.7.cross_attn.importance_net.2.weight torch.Size([1, 896])
380
- model.layers.7.cross_attn.importance_net.2.bias torch.Size([1])
381
- model.layers.8.self_attn.q_proj.weight torch.Size([3584, 3584])
382
- model.layers.8.self_attn.q_proj.bias torch.Size([3584])
383
- model.layers.8.self_attn.k_proj.weight torch.Size([512, 3584])
384
- model.layers.8.self_attn.k_proj.bias torch.Size([512])
385
- model.layers.8.self_attn.v_proj.weight torch.Size([512, 3584])
386
- model.layers.8.self_attn.v_proj.bias torch.Size([512])
387
- model.layers.8.self_attn.o_proj.weight torch.Size([3584, 3584])
388
- model.layers.8.mlp.gate_proj.weight torch.Size([18944, 3584])
389
- model.layers.8.mlp.up_proj.weight torch.Size([18944, 3584])
390
- model.layers.8.mlp.down_proj.weight torch.Size([3584, 18944])
391
- model.layers.8.input_layernorm.weight torch.Size([3584])
392
- model.layers.8.post_attention_layernorm.weight torch.Size([3584])
393
- model.layers.8.cross_attn.gate torch.Size([])
394
- model.layers.8.cross_attn.spatial_scale_param torch.Size([])
395
- model.layers.8.cross_attn.input_layernorm.weight torch.Size([3584])
396
- model.layers.8.cross_attn.input_layernorm.bias torch.Size([3584])
397
- model.layers.8.cross_attn.vggt_layernorm.weight torch.Size([3584])
398
- model.layers.8.cross_attn.vggt_layernorm.bias torch.Size([3584])
399
- model.layers.8.cross_attn.q_proj.weight torch.Size([3584, 3584])
400
- model.layers.8.cross_attn.k_proj.weight torch.Size([3584, 3584])
401
- model.layers.8.cross_attn.v_proj.weight torch.Size([3584, 3584])
402
- model.layers.8.cross_attn.o_proj.weight torch.Size([3584, 3584])
403
- model.layers.8.cross_attn.importance_net.0.weight torch.Size([896, 3584])
404
- model.layers.8.cross_attn.importance_net.0.bias torch.Size([896])
405
- model.layers.8.cross_attn.importance_net.2.weight torch.Size([1, 896])
406
- model.layers.8.cross_attn.importance_net.2.bias torch.Size([1])
407
- model.layers.9.self_attn.q_proj.weight torch.Size([3584, 3584])
408
- model.layers.9.self_attn.q_proj.bias torch.Size([3584])
409
- model.layers.9.self_attn.k_proj.weight torch.Size([512, 3584])
410
- model.layers.9.self_attn.k_proj.bias torch.Size([512])
411
- model.layers.9.self_attn.v_proj.weight torch.Size([512, 3584])
412
- model.layers.9.self_attn.v_proj.bias torch.Size([512])
413
- model.layers.9.self_attn.o_proj.weight torch.Size([3584, 3584])
414
- model.layers.9.mlp.gate_proj.weight torch.Size([18944, 3584])
415
- model.layers.9.mlp.up_proj.weight torch.Size([18944, 3584])
416
- model.layers.9.mlp.down_proj.weight torch.Size([3584, 18944])
417
- model.layers.9.input_layernorm.weight torch.Size([3584])
418
- model.layers.9.post_attention_layernorm.weight torch.Size([3584])
419
- model.layers.9.cross_attn.gate torch.Size([])
420
- model.layers.9.cross_attn.spatial_scale_param torch.Size([])
421
- model.layers.9.cross_attn.input_layernorm.weight torch.Size([3584])
422
- model.layers.9.cross_attn.input_layernorm.bias torch.Size([3584])
423
- model.layers.9.cross_attn.vggt_layernorm.weight torch.Size([3584])
424
- model.layers.9.cross_attn.vggt_layernorm.bias torch.Size([3584])
425
- model.layers.9.cross_attn.q_proj.weight torch.Size([3584, 3584])
426
- model.layers.9.cross_attn.k_proj.weight torch.Size([3584, 3584])
427
- model.layers.9.cross_attn.v_proj.weight torch.Size([3584, 3584])
428
- model.layers.9.cross_attn.o_proj.weight torch.Size([3584, 3584])
429
- model.layers.9.cross_attn.importance_net.0.weight torch.Size([896, 3584])
430
- model.layers.9.cross_attn.importance_net.0.bias torch.Size([896])
431
- model.layers.9.cross_attn.importance_net.2.weight torch.Size([1, 896])
432
- model.layers.9.cross_attn.importance_net.2.bias torch.Size([1])
433
- model.layers.10.self_attn.q_proj.weight torch.Size([3584, 3584])
434
- model.layers.10.self_attn.q_proj.bias torch.Size([3584])
435
- model.layers.10.self_attn.k_proj.weight torch.Size([512, 3584])
436
- model.layers.10.self_attn.k_proj.bias torch.Size([512])
437
- model.layers.10.self_attn.v_proj.weight torch.Size([512, 3584])
438
- model.layers.10.self_attn.v_proj.bias torch.Size([512])
439
- model.layers.10.self_attn.o_proj.weight torch.Size([3584, 3584])
440
- model.layers.10.mlp.gate_proj.weight torch.Size([18944, 3584])
441
- model.layers.10.mlp.up_proj.weight torch.Size([18944, 3584])
442
- model.layers.10.mlp.down_proj.weight torch.Size([3584, 18944])
443
- model.layers.10.input_layernorm.weight torch.Size([3584])
444
- model.layers.10.post_attention_layernorm.weight torch.Size([3584])
445
- model.layers.10.cross_attn.gate torch.Size([])
446
- model.layers.10.cross_attn.spatial_scale_param torch.Size([])
447
- model.layers.10.cross_attn.input_layernorm.weight torch.Size([3584])
448
- model.layers.10.cross_attn.input_layernorm.bias torch.Size([3584])
449
- model.layers.10.cross_attn.vggt_layernorm.weight torch.Size([3584])
450
- model.layers.10.cross_attn.vggt_layernorm.bias torch.Size([3584])
451
- model.layers.10.cross_attn.q_proj.weight torch.Size([3584, 3584])
452
- model.layers.10.cross_attn.k_proj.weight torch.Size([3584, 3584])
453
- model.layers.10.cross_attn.v_proj.weight torch.Size([3584, 3584])
454
- model.layers.10.cross_attn.o_proj.weight torch.Size([3584, 3584])
455
- model.layers.10.cross_attn.importance_net.0.weight torch.Size([896, 3584])
456
- model.layers.10.cross_attn.importance_net.0.bias torch.Size([896])
457
- model.layers.10.cross_attn.importance_net.2.weight torch.Size([1, 896])
458
- model.layers.10.cross_attn.importance_net.2.bias torch.Size([1])
459
- model.layers.11.self_attn.q_proj.weight torch.Size([3584, 3584])
460
- model.layers.11.self_attn.q_proj.bias torch.Size([3584])
461
- model.layers.11.self_attn.k_proj.weight torch.Size([512, 3584])
462
- model.layers.11.self_attn.k_proj.bias torch.Size([512])
463
- model.layers.11.self_attn.v_proj.weight torch.Size([512, 3584])
464
- model.layers.11.self_attn.v_proj.bias torch.Size([512])
465
- model.layers.11.self_attn.o_proj.weight torch.Size([3584, 3584])
466
- model.layers.11.mlp.gate_proj.weight torch.Size([18944, 3584])
467
- model.layers.11.mlp.up_proj.weight torch.Size([18944, 3584])
468
- model.layers.11.mlp.down_proj.weight torch.Size([3584, 18944])
469
- model.layers.11.input_layernorm.weight torch.Size([3584])
470
- model.layers.11.post_attention_layernorm.weight torch.Size([3584])
471
- model.layers.11.cross_attn.gate torch.Size([])
472
- model.layers.11.cross_attn.spatial_scale_param torch.Size([])
473
- model.layers.11.cross_attn.input_layernorm.weight torch.Size([3584])
474
- model.layers.11.cross_attn.input_layernorm.bias torch.Size([3584])
475
- model.layers.11.cross_attn.vggt_layernorm.weight torch.Size([3584])
476
- model.layers.11.cross_attn.vggt_layernorm.bias torch.Size([3584])
477
- model.layers.11.cross_attn.q_proj.weight torch.Size([3584, 3584])
478
- model.layers.11.cross_attn.k_proj.weight torch.Size([3584, 3584])
479
- model.layers.11.cross_attn.v_proj.weight torch.Size([3584, 3584])
480
- model.layers.11.cross_attn.o_proj.weight torch.Size([3584, 3584])
481
- model.layers.11.cross_attn.importance_net.0.weight torch.Size([896, 3584])
482
- model.layers.11.cross_attn.importance_net.0.bias torch.Size([896])
483
- model.layers.11.cross_attn.importance_net.2.weight torch.Size([1, 896])
484
- model.layers.11.cross_attn.importance_net.2.bias torch.Size([1])
485
- model.layers.12.self_attn.q_proj.weight torch.Size([3584, 3584])
486
- model.layers.12.self_attn.q_proj.bias torch.Size([3584])
487
- model.layers.12.self_attn.k_proj.weight torch.Size([512, 3584])
488
- model.layers.12.self_attn.k_proj.bias torch.Size([512])
489
- model.layers.12.self_attn.v_proj.weight torch.Size([512, 3584])
490
- model.layers.12.self_attn.v_proj.bias torch.Size([512])
491
- model.layers.12.self_attn.o_proj.weight torch.Size([3584, 3584])
492
- model.layers.12.mlp.gate_proj.weight torch.Size([18944, 3584])
493
- model.layers.12.mlp.up_proj.weight torch.Size([18944, 3584])
494
- model.layers.12.mlp.down_proj.weight torch.Size([3584, 18944])
495
- model.layers.12.input_layernorm.weight torch.Size([3584])
496
- model.layers.12.post_attention_layernorm.weight torch.Size([3584])
497
- model.layers.12.cross_attn.gate torch.Size([])
498
- model.layers.12.cross_attn.spatial_scale_param torch.Size([])
499
- model.layers.12.cross_attn.input_layernorm.weight torch.Size([3584])
500
- model.layers.12.cross_attn.input_layernorm.bias torch.Size([3584])
501
- model.layers.12.cross_attn.vggt_layernorm.weight torch.Size([3584])
502
- model.layers.12.cross_attn.vggt_layernorm.bias torch.Size([3584])
503
- model.layers.12.cross_attn.q_proj.weight torch.Size([3584, 3584])
504
- model.layers.12.cross_attn.k_proj.weight torch.Size([3584, 3584])
505
- model.layers.12.cross_attn.v_proj.weight torch.Size([3584, 3584])
506
- model.layers.12.cross_attn.o_proj.weight torch.Size([3584, 3584])
507
- model.layers.12.cross_attn.importance_net.0.weight torch.Size([896, 3584])
508
- model.layers.12.cross_attn.importance_net.0.bias torch.Size([896])
509
- model.layers.12.cross_attn.importance_net.2.weight torch.Size([1, 896])
510
- model.layers.12.cross_attn.importance_net.2.bias torch.Size([1])
511
- model.layers.13.self_attn.q_proj.weight torch.Size([3584, 3584])
512
- model.layers.13.self_attn.q_proj.bias torch.Size([3584])
513
- model.layers.13.self_attn.k_proj.weight torch.Size([512, 3584])
514
- model.layers.13.self_attn.k_proj.bias torch.Size([512])
515
- model.layers.13.self_attn.v_proj.weight torch.Size([512, 3584])
516
- model.layers.13.self_attn.v_proj.bias torch.Size([512])
517
- model.layers.13.self_attn.o_proj.weight torch.Size([3584, 3584])
518
- model.layers.13.mlp.gate_proj.weight torch.Size([18944, 3584])
519
- model.layers.13.mlp.up_proj.weight torch.Size([18944, 3584])
520
- model.layers.13.mlp.down_proj.weight torch.Size([3584, 18944])
521
- model.layers.13.input_layernorm.weight torch.Size([3584])
522
- model.layers.13.post_attention_layernorm.weight torch.Size([3584])
523
- model.layers.13.cross_attn.gate torch.Size([])
524
- model.layers.13.cross_attn.spatial_scale_param torch.Size([])
525
- model.layers.13.cross_attn.input_layernorm.weight torch.Size([3584])
526
- model.layers.13.cross_attn.input_layernorm.bias torch.Size([3584])
527
- model.layers.13.cross_attn.vggt_layernorm.weight torch.Size([3584])
528
- model.layers.13.cross_attn.vggt_layernorm.bias torch.Size([3584])
529
- model.layers.13.cross_attn.q_proj.weight torch.Size([3584, 3584])
530
- model.layers.13.cross_attn.k_proj.weight torch.Size([3584, 3584])
531
- model.layers.13.cross_attn.v_proj.weight torch.Size([3584, 3584])
532
- model.layers.13.cross_attn.o_proj.weight torch.Size([3584, 3584])
533
- model.layers.13.cross_attn.importance_net.0.weight torch.Size([896, 3584])
534
- model.layers.13.cross_attn.importance_net.0.bias torch.Size([896])
535
- model.layers.13.cross_attn.importance_net.2.weight torch.Size([1, 896])
536
- model.layers.13.cross_attn.importance_net.2.bias torch.Size([1])
537
- model.layers.14.self_attn.q_proj.weight torch.Size([3584, 3584])
538
- model.layers.14.self_attn.q_proj.bias torch.Size([3584])
539
- model.layers.14.self_attn.k_proj.weight torch.Size([512, 3584])
540
- model.layers.14.self_attn.k_proj.bias torch.Size([512])
541
- model.layers.14.self_attn.v_proj.weight torch.Size([512, 3584])
542
- model.layers.14.self_attn.v_proj.bias torch.Size([512])
543
- model.layers.14.self_attn.o_proj.weight torch.Size([3584, 3584])
544
- model.layers.14.mlp.gate_proj.weight torch.Size([18944, 3584])
545
- model.layers.14.mlp.up_proj.weight torch.Size([18944, 3584])
546
- model.layers.14.mlp.down_proj.weight torch.Size([3584, 18944])
547
- model.layers.14.input_layernorm.weight torch.Size([3584])
548
- model.layers.14.post_attention_layernorm.weight torch.Size([3584])
549
- model.layers.14.cross_attn.gate torch.Size([])
550
- model.layers.14.cross_attn.spatial_scale_param torch.Size([])
551
- model.layers.14.cross_attn.input_layernorm.weight torch.Size([3584])
552
- model.layers.14.cross_attn.input_layernorm.bias torch.Size([3584])
553
- model.layers.14.cross_attn.vggt_layernorm.weight torch.Size([3584])
554
- model.layers.14.cross_attn.vggt_layernorm.bias torch.Size([3584])
555
- model.layers.14.cross_attn.q_proj.weight torch.Size([3584, 3584])
556
- model.layers.14.cross_attn.k_proj.weight torch.Size([3584, 3584])
557
- model.layers.14.cross_attn.v_proj.weight torch.Size([3584, 3584])
558
- model.layers.14.cross_attn.o_proj.weight torch.Size([3584, 3584])
559
- model.layers.14.cross_attn.importance_net.0.weight torch.Size([896, 3584])
560
- model.layers.14.cross_attn.importance_net.0.bias torch.Size([896])
561
- model.layers.14.cross_attn.importance_net.2.weight torch.Size([1, 896])
562
- model.layers.14.cross_attn.importance_net.2.bias torch.Size([1])
563
- model.layers.15.self_attn.q_proj.weight torch.Size([3584, 3584])
564
- model.layers.15.self_attn.q_proj.bias torch.Size([3584])
565
- model.layers.15.self_attn.k_proj.weight torch.Size([512, 3584])
566
- model.layers.15.self_attn.k_proj.bias torch.Size([512])
567
- model.layers.15.self_attn.v_proj.weight torch.Size([512, 3584])
568
- model.layers.15.self_attn.v_proj.bias torch.Size([512])
569
- model.layers.15.self_attn.o_proj.weight torch.Size([3584, 3584])
570
- model.layers.15.mlp.gate_proj.weight torch.Size([18944, 3584])
571
- model.layers.15.mlp.up_proj.weight torch.Size([18944, 3584])
572
- model.layers.15.mlp.down_proj.weight torch.Size([3584, 18944])
573
- model.layers.15.input_layernorm.weight torch.Size([3584])
574
- model.layers.15.post_attention_layernorm.weight torch.Size([3584])
575
- model.layers.15.cross_attn.gate torch.Size([])
576
- model.layers.15.cross_attn.spatial_scale_param torch.Size([])
577
- model.layers.15.cross_attn.input_layernorm.weight torch.Size([3584])
578
- model.layers.15.cross_attn.input_layernorm.bias torch.Size([3584])
579
- model.layers.15.cross_attn.vggt_layernorm.weight torch.Size([3584])
580
- model.layers.15.cross_attn.vggt_layernorm.bias torch.Size([3584])
581
- model.layers.15.cross_attn.q_proj.weight torch.Size([3584, 3584])
582
- model.layers.15.cross_attn.k_proj.weight torch.Size([3584, 3584])
583
- model.layers.15.cross_attn.v_proj.weight torch.Size([3584, 3584])
584
- model.layers.15.cross_attn.o_proj.weight torch.Size([3584, 3584])
585
- model.layers.15.cross_attn.importance_net.0.weight torch.Size([896, 3584])
586
- model.layers.15.cross_attn.importance_net.0.bias torch.Size([896])
587
- model.layers.15.cross_attn.importance_net.2.weight torch.Size([1, 896])
588
- model.layers.15.cross_attn.importance_net.2.bias torch.Size([1])
589
- model.layers.16.self_attn.q_proj.weight torch.Size([3584, 3584])
590
- model.layers.16.self_attn.q_proj.bias torch.Size([3584])
591
- model.layers.16.self_attn.k_proj.weight torch.Size([512, 3584])
592
- model.layers.16.self_attn.k_proj.bias torch.Size([512])
593
- model.layers.16.self_attn.v_proj.weight torch.Size([512, 3584])
594
- model.layers.16.self_attn.v_proj.bias torch.Size([512])
595
- model.layers.16.self_attn.o_proj.weight torch.Size([3584, 3584])
596
- model.layers.16.mlp.gate_proj.weight torch.Size([18944, 3584])
597
- model.layers.16.mlp.up_proj.weight torch.Size([18944, 3584])
598
- model.layers.16.mlp.down_proj.weight torch.Size([3584, 18944])
599
- model.layers.16.input_layernorm.weight torch.Size([3584])
600
- model.layers.16.post_attention_layernorm.weight torch.Size([3584])
601
- model.layers.16.cross_attn.gate torch.Size([])
602
- model.layers.16.cross_attn.spatial_scale_param torch.Size([])
603
- model.layers.16.cross_attn.input_layernorm.weight torch.Size([3584])
604
- model.layers.16.cross_attn.input_layernorm.bias torch.Size([3584])
605
- model.layers.16.cross_attn.vggt_layernorm.weight torch.Size([3584])
606
- model.layers.16.cross_attn.vggt_layernorm.bias torch.Size([3584])
607
- model.layers.16.cross_attn.q_proj.weight torch.Size([3584, 3584])
608
- model.layers.16.cross_attn.k_proj.weight torch.Size([3584, 3584])
609
- model.layers.16.cross_attn.v_proj.weight torch.Size([3584, 3584])
610
- model.layers.16.cross_attn.o_proj.weight torch.Size([3584, 3584])
611
- model.layers.16.cross_attn.importance_net.0.weight torch.Size([896, 3584])
612
- model.layers.16.cross_attn.importance_net.0.bias torch.Size([896])
613
- model.layers.16.cross_attn.importance_net.2.weight torch.Size([1, 896])
614
- model.layers.16.cross_attn.importance_net.2.bias torch.Size([1])
615
- model.layers.17.self_attn.q_proj.weight torch.Size([3584, 3584])
616
- model.layers.17.self_attn.q_proj.bias torch.Size([3584])
617
- model.layers.17.self_attn.k_proj.weight torch.Size([512, 3584])
618
- model.layers.17.self_attn.k_proj.bias torch.Size([512])
619
- model.layers.17.self_attn.v_proj.weight torch.Size([512, 3584])
620
- model.layers.17.self_attn.v_proj.bias torch.Size([512])
621
- model.layers.17.self_attn.o_proj.weight torch.Size([3584, 3584])
622
- model.layers.17.mlp.gate_proj.weight torch.Size([18944, 3584])
623
- model.layers.17.mlp.up_proj.weight torch.Size([18944, 3584])
624
- model.layers.17.mlp.down_proj.weight torch.Size([3584, 18944])
625
- model.layers.17.input_layernorm.weight torch.Size([3584])
626
- model.layers.17.post_attention_layernorm.weight torch.Size([3584])
627
- model.layers.17.cross_attn.gate torch.Size([])
628
- model.layers.17.cross_attn.spatial_scale_param torch.Size([])
629
- model.layers.17.cross_attn.input_layernorm.weight torch.Size([3584])
630
- model.layers.17.cross_attn.input_layernorm.bias torch.Size([3584])
631
- model.layers.17.cross_attn.vggt_layernorm.weight torch.Size([3584])
632
- model.layers.17.cross_attn.vggt_layernorm.bias torch.Size([3584])
633
- model.layers.17.cross_attn.q_proj.weight torch.Size([3584, 3584])
634
- model.layers.17.cross_attn.k_proj.weight torch.Size([3584, 3584])
635
- model.layers.17.cross_attn.v_proj.weight torch.Size([3584, 3584])
636
- model.layers.17.cross_attn.o_proj.weight torch.Size([3584, 3584])
637
- model.layers.17.cross_attn.importance_net.0.weight torch.Size([896, 3584])
638
- model.layers.17.cross_attn.importance_net.0.bias torch.Size([896])
639
- model.layers.17.cross_attn.importance_net.2.weight torch.Size([1, 896])
640
- model.layers.17.cross_attn.importance_net.2.bias torch.Size([1])
641
- model.layers.18.self_attn.q_proj.weight torch.Size([3584, 3584])
642
- model.layers.18.self_attn.q_proj.bias torch.Size([3584])
643
- model.layers.18.self_attn.k_proj.weight torch.Size([512, 3584])
644
- model.layers.18.self_attn.k_proj.bias torch.Size([512])
645
- model.layers.18.self_attn.v_proj.weight torch.Size([512, 3584])
646
- model.layers.18.self_attn.v_proj.bias torch.Size([512])
647
- model.layers.18.self_attn.o_proj.weight torch.Size([3584, 3584])
648
- model.layers.18.mlp.gate_proj.weight torch.Size([18944, 3584])
649
- model.layers.18.mlp.up_proj.weight torch.Size([18944, 3584])
650
- model.layers.18.mlp.down_proj.weight torch.Size([3584, 18944])
651
- model.layers.18.input_layernorm.weight torch.Size([3584])
652
- model.layers.18.post_attention_layernorm.weight torch.Size([3584])
653
- model.layers.18.cross_attn.gate torch.Size([])
654
- model.layers.18.cross_attn.spatial_scale_param torch.Size([])
655
- model.layers.18.cross_attn.input_layernorm.weight torch.Size([3584])
656
- model.layers.18.cross_attn.input_layernorm.bias torch.Size([3584])
657
- model.layers.18.cross_attn.vggt_layernorm.weight torch.Size([3584])
658
- model.layers.18.cross_attn.vggt_layernorm.bias torch.Size([3584])
659
- model.layers.18.cross_attn.q_proj.weight torch.Size([3584, 3584])
660
- model.layers.18.cross_attn.k_proj.weight torch.Size([3584, 3584])
661
- model.layers.18.cross_attn.v_proj.weight torch.Size([3584, 3584])
662
- model.layers.18.cross_attn.o_proj.weight torch.Size([3584, 3584])
663
- model.layers.18.cross_attn.importance_net.0.weight torch.Size([896, 3584])
664
- model.layers.18.cross_attn.importance_net.0.bias torch.Size([896])
665
- model.layers.18.cross_attn.importance_net.2.weight torch.Size([1, 896])
666
- model.layers.18.cross_attn.importance_net.2.bias torch.Size([1])
667
- model.layers.19.self_attn.q_proj.weight torch.Size([3584, 3584])
668
- model.layers.19.self_attn.q_proj.bias torch.Size([3584])
669
- model.layers.19.self_attn.k_proj.weight torch.Size([512, 3584])
670
- model.layers.19.self_attn.k_proj.bias torch.Size([512])
671
- model.layers.19.self_attn.v_proj.weight torch.Size([512, 3584])
672
- model.layers.19.self_attn.v_proj.bias torch.Size([512])
673
- model.layers.19.self_attn.o_proj.weight torch.Size([3584, 3584])
674
- model.layers.19.mlp.gate_proj.weight torch.Size([18944, 3584])
675
- model.layers.19.mlp.up_proj.weight torch.Size([18944, 3584])
676
- model.layers.19.mlp.down_proj.weight torch.Size([3584, 18944])
677
- model.layers.19.input_layernorm.weight torch.Size([3584])
678
- model.layers.19.post_attention_layernorm.weight torch.Size([3584])
679
- model.layers.19.cross_attn.gate torch.Size([])
680
- model.layers.19.cross_attn.spatial_scale_param torch.Size([])
681
- model.layers.19.cross_attn.input_layernorm.weight torch.Size([3584])
682
- model.layers.19.cross_attn.input_layernorm.bias torch.Size([3584])
683
- model.layers.19.cross_attn.vggt_layernorm.weight torch.Size([3584])
684
- model.layers.19.cross_attn.vggt_layernorm.bias torch.Size([3584])
685
- model.layers.19.cross_attn.q_proj.weight torch.Size([3584, 3584])
686
- model.layers.19.cross_attn.k_proj.weight torch.Size([3584, 3584])
687
- model.layers.19.cross_attn.v_proj.weight torch.Size([3584, 3584])
688
- model.layers.19.cross_attn.o_proj.weight torch.Size([3584, 3584])
689
- model.layers.19.cross_attn.importance_net.0.weight torch.Size([896, 3584])
690
- model.layers.19.cross_attn.importance_net.0.bias torch.Size([896])
691
- model.layers.19.cross_attn.importance_net.2.weight torch.Size([1, 896])
692
- model.layers.19.cross_attn.importance_net.2.bias torch.Size([1])
693
- model.layers.20.self_attn.q_proj.weight torch.Size([3584, 3584])
694
- model.layers.20.self_attn.q_proj.bias torch.Size([3584])
695
- model.layers.20.self_attn.k_proj.weight torch.Size([512, 3584])
696
- model.layers.20.self_attn.k_proj.bias torch.Size([512])
697
- model.layers.20.self_attn.v_proj.weight torch.Size([512, 3584])
698
- model.layers.20.self_attn.v_proj.bias torch.Size([512])
699
- model.layers.20.self_attn.o_proj.weight torch.Size([3584, 3584])
700
- model.layers.20.mlp.gate_proj.weight torch.Size([18944, 3584])
701
- model.layers.20.mlp.up_proj.weight torch.Size([18944, 3584])
702
- model.layers.20.mlp.down_proj.weight torch.Size([3584, 18944])
703
- model.layers.20.input_layernorm.weight torch.Size([3584])
704
- model.layers.20.post_attention_layernorm.weight torch.Size([3584])
705
- model.layers.20.cross_attn.gate torch.Size([])
706
- model.layers.20.cross_attn.spatial_scale_param torch.Size([])
707
- model.layers.20.cross_attn.input_layernorm.weight torch.Size([3584])
708
- model.layers.20.cross_attn.input_layernorm.bias torch.Size([3584])
709
- model.layers.20.cross_attn.vggt_layernorm.weight torch.Size([3584])
710
- model.layers.20.cross_attn.vggt_layernorm.bias torch.Size([3584])
711
- model.layers.20.cross_attn.q_proj.weight torch.Size([3584, 3584])
712
- model.layers.20.cross_attn.k_proj.weight torch.Size([3584, 3584])
713
- model.layers.20.cross_attn.v_proj.weight torch.Size([3584, 3584])
714
- model.layers.20.cross_attn.o_proj.weight torch.Size([3584, 3584])
715
- model.layers.20.cross_attn.importance_net.0.weight torch.Size([896, 3584])
716
- model.layers.20.cross_attn.importance_net.0.bias torch.Size([896])
717
- model.layers.20.cross_attn.importance_net.2.weight torch.Size([1, 896])
718
- model.layers.20.cross_attn.importance_net.2.bias torch.Size([1])
719
- model.layers.21.self_attn.q_proj.weight torch.Size([3584, 3584])
720
- model.layers.21.self_attn.q_proj.bias torch.Size([3584])
721
- model.layers.21.self_attn.k_proj.weight torch.Size([512, 3584])
722
- model.layers.21.self_attn.k_proj.bias torch.Size([512])
723
- model.layers.21.self_attn.v_proj.weight torch.Size([512, 3584])
724
- model.layers.21.self_attn.v_proj.bias torch.Size([512])
725
- model.layers.21.self_attn.o_proj.weight torch.Size([3584, 3584])
726
- model.layers.21.mlp.gate_proj.weight torch.Size([18944, 3584])
727
- model.layers.21.mlp.up_proj.weight torch.Size([18944, 3584])
728
- model.layers.21.mlp.down_proj.weight torch.Size([3584, 18944])
729
- model.layers.21.input_layernorm.weight torch.Size([3584])
730
- model.layers.21.post_attention_layernorm.weight torch.Size([3584])
731
- model.layers.21.cross_attn.gate torch.Size([])
732
- model.layers.21.cross_attn.spatial_scale_param torch.Size([])
733
- model.layers.21.cross_attn.input_layernorm.weight torch.Size([3584])
734
- model.layers.21.cross_attn.input_layernorm.bias torch.Size([3584])
735
- model.layers.21.cross_attn.vggt_layernorm.weight torch.Size([3584])
736
- model.layers.21.cross_attn.vggt_layernorm.bias torch.Size([3584])
737
- model.layers.21.cross_attn.q_proj.weight torch.Size([3584, 3584])
738
- model.layers.21.cross_attn.k_proj.weight torch.Size([3584, 3584])
739
- model.layers.21.cross_attn.v_proj.weight torch.Size([3584, 3584])
740
- model.layers.21.cross_attn.o_proj.weight torch.Size([3584, 3584])
741
- model.layers.21.cross_attn.importance_net.0.weight torch.Size([896, 3584])
742
- model.layers.21.cross_attn.importance_net.0.bias torch.Size([896])
743
- model.layers.21.cross_attn.importance_net.2.weight torch.Size([1, 896])
744
- model.layers.21.cross_attn.importance_net.2.bias torch.Size([1])
745
- model.layers.22.self_attn.q_proj.weight torch.Size([3584, 3584])
746
- model.layers.22.self_attn.q_proj.bias torch.Size([3584])
747
- model.layers.22.self_attn.k_proj.weight torch.Size([512, 3584])
748
- model.layers.22.self_attn.k_proj.bias torch.Size([512])
749
- model.layers.22.self_attn.v_proj.weight torch.Size([512, 3584])
750
- model.layers.22.self_attn.v_proj.bias torch.Size([512])
751
- model.layers.22.self_attn.o_proj.weight torch.Size([3584, 3584])
752
- model.layers.22.mlp.gate_proj.weight torch.Size([18944, 3584])
753
- model.layers.22.mlp.up_proj.weight torch.Size([18944, 3584])
754
- model.layers.22.mlp.down_proj.weight torch.Size([3584, 18944])
755
- model.layers.22.input_layernorm.weight torch.Size([3584])
756
- model.layers.22.post_attention_layernorm.weight torch.Size([3584])
757
- model.layers.22.cross_attn.gate torch.Size([])
758
- model.layers.22.cross_attn.spatial_scale_param torch.Size([])
759
- model.layers.22.cross_attn.input_layernorm.weight torch.Size([3584])
760
- model.layers.22.cross_attn.input_layernorm.bias torch.Size([3584])
761
- model.layers.22.cross_attn.vggt_layernorm.weight torch.Size([3584])
762
- model.layers.22.cross_attn.vggt_layernorm.bias torch.Size([3584])
763
- model.layers.22.cross_attn.q_proj.weight torch.Size([3584, 3584])
764
- model.layers.22.cross_attn.k_proj.weight torch.Size([3584, 3584])
765
- model.layers.22.cross_attn.v_proj.weight torch.Size([3584, 3584])
766
- model.layers.22.cross_attn.o_proj.weight torch.Size([3584, 3584])
767
- model.layers.22.cross_attn.importance_net.0.weight torch.Size([896, 3584])
768
- model.layers.22.cross_attn.importance_net.0.bias torch.Size([896])
769
- model.layers.22.cross_attn.importance_net.2.weight torch.Size([1, 896])
770
- model.layers.22.cross_attn.importance_net.2.bias torch.Size([1])
771
- model.layers.23.self_attn.q_proj.weight torch.Size([3584, 3584])
772
- model.layers.23.self_attn.q_proj.bias torch.Size([3584])
773
- model.layers.23.self_attn.k_proj.weight torch.Size([512, 3584])
774
- model.layers.23.self_attn.k_proj.bias torch.Size([512])
775
- model.layers.23.self_attn.v_proj.weight torch.Size([512, 3584])
776
- model.layers.23.self_attn.v_proj.bias torch.Size([512])
777
- model.layers.23.self_attn.o_proj.weight torch.Size([3584, 3584])
778
- model.layers.23.mlp.gate_proj.weight torch.Size([18944, 3584])
779
- model.layers.23.mlp.up_proj.weight torch.Size([18944, 3584])
780
- model.layers.23.mlp.down_proj.weight torch.Size([3584, 18944])
781
- model.layers.23.input_layernorm.weight torch.Size([3584])
782
- model.layers.23.post_attention_layernorm.weight torch.Size([3584])
783
- model.layers.23.cross_attn.gate torch.Size([])
784
- model.layers.23.cross_attn.spatial_scale_param torch.Size([])
785
- model.layers.23.cross_attn.input_layernorm.weight torch.Size([3584])
786
- model.layers.23.cross_attn.input_layernorm.bias torch.Size([3584])
787
- model.layers.23.cross_attn.vggt_layernorm.weight torch.Size([3584])
788
- model.layers.23.cross_attn.vggt_layernorm.bias torch.Size([3584])
789
- model.layers.23.cross_attn.q_proj.weight torch.Size([3584, 3584])
790
- model.layers.23.cross_attn.k_proj.weight torch.Size([3584, 3584])
791
- model.layers.23.cross_attn.v_proj.weight torch.Size([3584, 3584])
792
- model.layers.23.cross_attn.o_proj.weight torch.Size([3584, 3584])
793
- model.layers.23.cross_attn.importance_net.0.weight torch.Size([896, 3584])
794
- model.layers.23.cross_attn.importance_net.0.bias torch.Size([896])
795
- model.layers.23.cross_attn.importance_net.2.weight torch.Size([1, 896])
796
- model.layers.23.cross_attn.importance_net.2.bias torch.Size([1])
797
- model.layers.24.self_attn.q_proj.weight torch.Size([3584, 3584])
798
- model.layers.24.self_attn.q_proj.bias torch.Size([3584])
799
- model.layers.24.self_attn.k_proj.weight torch.Size([512, 3584])
800
- model.layers.24.self_attn.k_proj.bias torch.Size([512])
801
- model.layers.24.self_attn.v_proj.weight torch.Size([512, 3584])
802
- model.layers.24.self_attn.v_proj.bias torch.Size([512])
803
- model.layers.24.self_attn.o_proj.weight torch.Size([3584, 3584])
804
- model.layers.24.mlp.gate_proj.weight torch.Size([18944, 3584])
805
- model.layers.24.mlp.up_proj.weight torch.Size([18944, 3584])
806
- model.layers.24.mlp.down_proj.weight torch.Size([3584, 18944])
807
- model.layers.24.input_layernorm.weight torch.Size([3584])
808
- model.layers.24.post_attention_layernorm.weight torch.Size([3584])
809
- model.layers.24.cross_attn.gate torch.Size([])
810
- model.layers.24.cross_attn.spatial_scale_param torch.Size([])
811
- model.layers.24.cross_attn.input_layernorm.weight torch.Size([3584])
812
- model.layers.24.cross_attn.input_layernorm.bias torch.Size([3584])
813
- model.layers.24.cross_attn.vggt_layernorm.weight torch.Size([3584])
814
- model.layers.24.cross_attn.vggt_layernorm.bias torch.Size([3584])
815
- model.layers.24.cross_attn.q_proj.weight torch.Size([3584, 3584])
816
- model.layers.24.cross_attn.k_proj.weight torch.Size([3584, 3584])
817
- model.layers.24.cross_attn.v_proj.weight torch.Size([3584, 3584])
818
- model.layers.24.cross_attn.o_proj.weight torch.Size([3584, 3584])
819
- model.layers.24.cross_attn.importance_net.0.weight torch.Size([896, 3584])
820
- model.layers.24.cross_attn.importance_net.0.bias torch.Size([896])
821
- model.layers.24.cross_attn.importance_net.2.weight torch.Size([1, 896])
822
- model.layers.24.cross_attn.importance_net.2.bias torch.Size([1])
823
- model.layers.25.self_attn.q_proj.weight torch.Size([3584, 3584])
824
- model.layers.25.self_attn.q_proj.bias torch.Size([3584])
825
- model.layers.25.self_attn.k_proj.weight torch.Size([512, 3584])
826
- model.layers.25.self_attn.k_proj.bias torch.Size([512])
827
- model.layers.25.self_attn.v_proj.weight torch.Size([512, 3584])
828
- model.layers.25.self_attn.v_proj.bias torch.Size([512])
829
- model.layers.25.self_attn.o_proj.weight torch.Size([3584, 3584])
830
- model.layers.25.mlp.gate_proj.weight torch.Size([18944, 3584])
831
- model.layers.25.mlp.up_proj.weight torch.Size([18944, 3584])
832
- model.layers.25.mlp.down_proj.weight torch.Size([3584, 18944])
833
- model.layers.25.input_layernorm.weight torch.Size([3584])
834
- model.layers.25.post_attention_layernorm.weight torch.Size([3584])
835
- model.layers.25.cross_attn.gate torch.Size([])
836
- model.layers.25.cross_attn.spatial_scale_param torch.Size([])
837
- model.layers.25.cross_attn.input_layernorm.weight torch.Size([3584])
838
- model.layers.25.cross_attn.input_layernorm.bias torch.Size([3584])
839
- model.layers.25.cross_attn.vggt_layernorm.weight torch.Size([3584])
840
- model.layers.25.cross_attn.vggt_layernorm.bias torch.Size([3584])
841
- model.layers.25.cross_attn.q_proj.weight torch.Size([3584, 3584])
842
- model.layers.25.cross_attn.k_proj.weight torch.Size([3584, 3584])
843
- model.layers.25.cross_attn.v_proj.weight torch.Size([3584, 3584])
844
- model.layers.25.cross_attn.o_proj.weight torch.Size([3584, 3584])
845
- model.layers.25.cross_attn.importance_net.0.weight torch.Size([896, 3584])
846
- model.layers.25.cross_attn.importance_net.0.bias torch.Size([896])
847
- model.layers.25.cross_attn.importance_net.2.weight torch.Size([1, 896])
848
- model.layers.25.cross_attn.importance_net.2.bias torch.Size([1])
849
- model.layers.26.self_attn.q_proj.weight torch.Size([3584, 3584])
850
- model.layers.26.self_attn.q_proj.bias torch.Size([3584])
851
- model.layers.26.self_attn.k_proj.weight torch.Size([512, 3584])
852
- model.layers.26.self_attn.k_proj.bias torch.Size([512])
853
- model.layers.26.self_attn.v_proj.weight torch.Size([512, 3584])
854
- model.layers.26.self_attn.v_proj.bias torch.Size([512])
855
- model.layers.26.self_attn.o_proj.weight torch.Size([3584, 3584])
856
- model.layers.26.mlp.gate_proj.weight torch.Size([18944, 3584])
857
- model.layers.26.mlp.up_proj.weight torch.Size([18944, 3584])
858
- model.layers.26.mlp.down_proj.weight torch.Size([3584, 18944])
859
- model.layers.26.input_layernorm.weight torch.Size([3584])
860
- model.layers.26.post_attention_layernorm.weight torch.Size([3584])
861
- model.layers.26.cross_attn.gate torch.Size([])
862
- model.layers.26.cross_attn.spatial_scale_param torch.Size([])
863
- model.layers.26.cross_attn.input_layernorm.weight torch.Size([3584])
864
- model.layers.26.cross_attn.input_layernorm.bias torch.Size([3584])
865
- model.layers.26.cross_attn.vggt_layernorm.weight torch.Size([3584])
866
- model.layers.26.cross_attn.vggt_layernorm.bias torch.Size([3584])
867
- model.layers.26.cross_attn.q_proj.weight torch.Size([3584, 3584])
868
- model.layers.26.cross_attn.k_proj.weight torch.Size([3584, 3584])
869
- model.layers.26.cross_attn.v_proj.weight torch.Size([3584, 3584])
870
- model.layers.26.cross_attn.o_proj.weight torch.Size([3584, 3584])
871
- model.layers.26.cross_attn.importance_net.0.weight torch.Size([896, 3584])
872
- model.layers.26.cross_attn.importance_net.0.bias torch.Size([896])
873
- model.layers.26.cross_attn.importance_net.2.weight torch.Size([1, 896])
874
- model.layers.26.cross_attn.importance_net.2.bias torch.Size([1])
875
- model.layers.27.self_attn.q_proj.weight torch.Size([3584, 3584])
876
- model.layers.27.self_attn.q_proj.bias torch.Size([3584])
877
- model.layers.27.self_attn.k_proj.weight torch.Size([512, 3584])
878
- model.layers.27.self_attn.k_proj.bias torch.Size([512])
879
- model.layers.27.self_attn.v_proj.weight torch.Size([512, 3584])
880
- model.layers.27.self_attn.v_proj.bias torch.Size([512])
881
- model.layers.27.self_attn.o_proj.weight torch.Size([3584, 3584])
882
- model.layers.27.mlp.gate_proj.weight torch.Size([18944, 3584])
883
- model.layers.27.mlp.up_proj.weight torch.Size([18944, 3584])
884
- model.layers.27.mlp.down_proj.weight torch.Size([3584, 18944])
885
- model.layers.27.input_layernorm.weight torch.Size([3584])
886
- model.layers.27.post_attention_layernorm.weight torch.Size([3584])
887
- model.layers.27.cross_attn.gate torch.Size([])
888
- model.layers.27.cross_attn.spatial_scale_param torch.Size([])
889
- model.layers.27.cross_attn.input_layernorm.weight torch.Size([3584])
890
- model.layers.27.cross_attn.input_layernorm.bias torch.Size([3584])
891
- model.layers.27.cross_attn.vggt_layernorm.weight torch.Size([3584])
892
- model.layers.27.cross_attn.vggt_layernorm.bias torch.Size([3584])
893
- model.layers.27.cross_attn.q_proj.weight torch.Size([3584, 3584])
894
- model.layers.27.cross_attn.k_proj.weight torch.Size([3584, 3584])
895
- model.layers.27.cross_attn.v_proj.weight torch.Size([3584, 3584])
896
- model.layers.27.cross_attn.o_proj.weight torch.Size([3584, 3584])
897
- model.layers.27.cross_attn.importance_net.0.weight torch.Size([896, 3584])
898
- model.layers.27.cross_attn.importance_net.0.bias torch.Size([896])
899
- model.layers.27.cross_attn.importance_net.2.weight torch.Size([1, 896])
900
- model.layers.27.cross_attn.importance_net.2.bias torch.Size([1])
901
- model.norm.weight torch.Size([3584])
902
- lm_head.weight torch.Size([152064, 3584])
903
- dataset_names ['llava_hound_64k', 'spar_234k']
904
- Loading datasets: [{'annotation_path': 'data/train/llava_hound_64k.json', 'data_path': 'data/media', 'tag': '2d', 'sampling_rate': 1.0, 'dataset_name': 'llava_hound_64k'}, {'annotation_path': 'data/train/spar_234k.json', 'data_path': 'data/media', 'tag': '3d', 'sampling_rate': 1.0, 'dataset_name': 'spar_234k'}]
905
- dataset_names ['llava_hound_64k', 'spar_234k']
906
- Loading datasets: [{'annotation_path': 'data/train/llava_hound_64k.json', 'data_path': 'data/media', 'tag': '2d', 'sampling_rate': 1.0, 'dataset_name': 'llava_hound_64k'}, {'annotation_path': 'data/train/spar_234k.json', 'data_path': 'data/media', 'tag': '3d', 'sampling_rate': 1.0, 'dataset_name': 'spar_234k'}]
907
- dataset_names ['llava_hound_64k', 'spar_234k']
908
- Loading datasets: [{'annotation_path': 'data/train/llava_hound_64k.json', 'data_path': 'data/media', 'tag': '2d', 'sampling_rate': 1.0, 'dataset_name': 'llava_hound_64k'}, {'annotation_path': 'data/train/spar_234k.json', 'data_path': 'data/media', 'tag': '3d', 'sampling_rate': 1.0, 'dataset_name': 'spar_234k'}]
909
- dataset_names ['llava_hound_64k', 'spar_234k']
910
- Loading datasets: [{'annotation_path': 'data/train/llava_hound_64k.json', 'data_path': 'data/media', 'tag': '2d', 'sampling_rate': 1.0, 'dataset_name': 'llava_hound_64k'}, {'annotation_path': 'data/train/spar_234k.json', 'data_path': 'data/media', 'tag': '3d', 'sampling_rate': 1.0, 'dataset_name': 'spar_234k'}]
911
- dataset_names ['llava_hound_64k', 'spar_234k']
912
- Loading datasets: [{'annotation_path': 'data/train/llava_hound_64k.json', 'data_path': 'data/media', 'tag': '2d', 'sampling_rate': 1.0, 'dataset_name': 'llava_hound_64k'}, {'annotation_path': 'data/train/spar_234k.json', 'data_path': 'data/media', 'tag': '3d', 'sampling_rate': 1.0, 'dataset_name': 'spar_234k'}]
913
- Total training samples: 298027
914
- Formatting inputs...Skip in lazy mode
915
- Total training samples: 298027
916
- Total training samples: 298027
917
- Total training samples: 298027
918
- Total training samples: 298027
919
- Total training samples: 298027
920
- Formatting inputs...Skip in lazy mode
921
- Formatting inputs...Skip in lazy mode
922
- Formatting inputs...Skip in lazy mode
923
- Total training samples: 298027
924
- Total training samples: 298027
925
- Formatting inputs...Skip in lazy mode
926
- Formatting inputs...Skip in lazy mode
927
- Formatting inputs...Skip in lazy mode
928
- Formatting inputs...Skip in lazy mode
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/trainer_state.json DELETED
@@ -1,3298 +0,0 @@
1
- {
2
- "best_global_step": null,
3
- "best_metric": null,
4
- "best_model_checkpoint": null,
5
- "epoch": 0.9998389434691577,
6
- "eval_steps": 500,
7
- "global_step": 4656,
8
- "is_hyper_param_search": false,
9
- "is_local_process_zero": true,
10
- "is_world_process_zero": true,
11
- "log_history": [
12
- {
13
- "epoch": 0.0021474204112310086,
14
- "grad_norm": 114.95087432861328,
15
- "learning_rate": 7.142857142857143e-07,
16
- "loss": 8.1645,
17
- "step": 10
18
- },
19
- {
20
- "epoch": 0.004294840822462017,
21
- "grad_norm": 57.820430755615234,
22
- "learning_rate": 1.4285714285714286e-06,
23
- "loss": 7.1175,
24
- "step": 20
25
- },
26
- {
27
- "epoch": 0.006442261233693026,
28
- "grad_norm": 38.31983947753906,
29
- "learning_rate": 2.1428571428571427e-06,
30
- "loss": 6.2552,
31
- "step": 30
32
- },
33
- {
34
- "epoch": 0.008589681644924034,
35
- "grad_norm": 30.298812866210938,
36
- "learning_rate": 2.8571428571428573e-06,
37
- "loss": 5.4179,
38
- "step": 40
39
- },
40
- {
41
- "epoch": 0.010737102056155044,
42
- "grad_norm": 31.72263526916504,
43
- "learning_rate": 3.5714285714285718e-06,
44
- "loss": 4.9746,
45
- "step": 50
46
- },
47
- {
48
- "epoch": 0.012884522467386052,
49
- "grad_norm": 25.22931480407715,
50
- "learning_rate": 4.2857142857142855e-06,
51
- "loss": 4.6509,
52
- "step": 60
53
- },
54
- {
55
- "epoch": 0.01503194287861706,
56
- "grad_norm": 28.012941360473633,
57
- "learning_rate": 5e-06,
58
- "loss": 4.2574,
59
- "step": 70
60
- },
61
- {
62
- "epoch": 0.01717936328984807,
63
- "grad_norm": 27.468263626098633,
64
- "learning_rate": 5.7142857142857145e-06,
65
- "loss": 3.806,
66
- "step": 80
67
- },
68
- {
69
- "epoch": 0.01932678370107908,
70
- "grad_norm": 22.046875,
71
- "learning_rate": 6.4285714285714295e-06,
72
- "loss": 3.8487,
73
- "step": 90
74
- },
75
- {
76
- "epoch": 0.02147420411231009,
77
- "grad_norm": 20.30846405029297,
78
- "learning_rate": 7.1428571428571436e-06,
79
- "loss": 2.9113,
80
- "step": 100
81
- },
82
- {
83
- "epoch": 0.023621624523541097,
84
- "grad_norm": 23.86836051940918,
85
- "learning_rate": 7.857142857142858e-06,
86
- "loss": 3.472,
87
- "step": 110
88
- },
89
- {
90
- "epoch": 0.025769044934772105,
91
- "grad_norm": 26.69591522216797,
92
- "learning_rate": 8.571428571428571e-06,
93
- "loss": 3.6422,
94
- "step": 120
95
- },
96
- {
97
- "epoch": 0.027916465346003113,
98
- "grad_norm": 20.753564834594727,
99
- "learning_rate": 9.285714285714288e-06,
100
- "loss": 3.0168,
101
- "step": 130
102
- },
103
- {
104
- "epoch": 0.03006388575723412,
105
- "grad_norm": 29.848037719726562,
106
- "learning_rate": 1e-05,
107
- "loss": 3.9783,
108
- "step": 140
109
- },
110
- {
111
- "epoch": 0.03221130616846513,
112
- "grad_norm": 20.223854064941406,
113
- "learning_rate": 9.999879015387978e-06,
114
- "loss": 3.51,
115
- "step": 150
116
- },
117
- {
118
- "epoch": 0.03435872657969614,
119
- "grad_norm": 18.951417922973633,
120
- "learning_rate": 9.999516067406818e-06,
121
- "loss": 2.9719,
122
- "step": 160
123
- },
124
- {
125
- "epoch": 0.03650614699092715,
126
- "grad_norm": 18.97931671142578,
127
- "learning_rate": 9.998911173620972e-06,
128
- "loss": 2.2938,
129
- "step": 170
130
- },
131
- {
132
- "epoch": 0.03865356740215816,
133
- "grad_norm": 23.515886306762695,
134
- "learning_rate": 9.998064363303573e-06,
135
- "loss": 2.8837,
136
- "step": 180
137
- },
138
- {
139
- "epoch": 0.040800987813389165,
140
- "grad_norm": 22.68735694885254,
141
- "learning_rate": 9.996975677435031e-06,
142
- "loss": 2.8285,
143
- "step": 190
144
- },
145
- {
146
- "epoch": 0.04294840822462018,
147
- "grad_norm": 20.7940731048584,
148
- "learning_rate": 9.995645168701038e-06,
149
- "loss": 2.1357,
150
- "step": 200
151
- },
152
- {
153
- "epoch": 0.04509582863585118,
154
- "grad_norm": 26.744306564331055,
155
- "learning_rate": 9.994072901490028e-06,
156
- "loss": 2.807,
157
- "step": 210
158
- },
159
- {
160
- "epoch": 0.04724324904708219,
161
- "grad_norm": 18.151416778564453,
162
- "learning_rate": 9.992258951890057e-06,
163
- "loss": 2.0274,
164
- "step": 220
165
- },
166
- {
167
- "epoch": 0.0493906694583132,
168
- "grad_norm": 17.781078338623047,
169
- "learning_rate": 9.990203407685122e-06,
170
- "loss": 2.6628,
171
- "step": 230
172
- },
173
- {
174
- "epoch": 0.05153808986954421,
175
- "grad_norm": 23.66128158569336,
176
- "learning_rate": 9.987906368350908e-06,
177
- "loss": 3.2315,
178
- "step": 240
179
- },
180
- {
181
- "epoch": 0.05368551028077522,
182
- "grad_norm": 23.81083106994629,
183
- "learning_rate": 9.98536794504998e-06,
184
- "loss": 2.971,
185
- "step": 250
186
- },
187
- {
188
- "epoch": 0.055832930692006226,
189
- "grad_norm": 23.036523818969727,
190
- "learning_rate": 9.982588260626402e-06,
191
- "loss": 2.5181,
192
- "step": 260
193
- },
194
- {
195
- "epoch": 0.05798035110323724,
196
- "grad_norm": 15.273154258728027,
197
- "learning_rate": 9.979567449599793e-06,
198
- "loss": 2.542,
199
- "step": 270
200
- },
201
- {
202
- "epoch": 0.06012777151446824,
203
- "grad_norm": 14.187295913696289,
204
- "learning_rate": 9.976305658158806e-06,
205
- "loss": 3.2005,
206
- "step": 280
207
- },
208
- {
209
- "epoch": 0.062275191925699254,
210
- "grad_norm": 25.680112838745117,
211
- "learning_rate": 9.972803044154078e-06,
212
- "loss": 2.1621,
213
- "step": 290
214
- },
215
- {
216
- "epoch": 0.06442261233693027,
217
- "grad_norm": 24.12076759338379,
218
- "learning_rate": 9.969059777090564e-06,
219
- "loss": 3.1866,
220
- "step": 300
221
- },
222
- {
223
- "epoch": 0.06657003274816127,
224
- "grad_norm": 16.713743209838867,
225
- "learning_rate": 9.965076038119347e-06,
226
- "loss": 1.8246,
227
- "step": 310
228
- },
229
- {
230
- "epoch": 0.06871745315939227,
231
- "grad_norm": 23.78823471069336,
232
- "learning_rate": 9.960852020028877e-06,
233
- "loss": 3.2128,
234
- "step": 320
235
- },
236
- {
237
- "epoch": 0.0708648735706233,
238
- "grad_norm": 24.15093994140625,
239
- "learning_rate": 9.956387927235628e-06,
240
- "loss": 2.8004,
241
- "step": 330
242
- },
243
- {
244
- "epoch": 0.0730122939818543,
245
- "grad_norm": 23.038162231445312,
246
- "learning_rate": 9.951683975774213e-06,
247
- "loss": 2.3707,
248
- "step": 340
249
- },
250
- {
251
- "epoch": 0.0751597143930853,
252
- "grad_norm": 24.0445556640625,
253
- "learning_rate": 9.946740393286928e-06,
254
- "loss": 2.3701,
255
- "step": 350
256
- },
257
- {
258
- "epoch": 0.07730713480431632,
259
- "grad_norm": 23.56902313232422,
260
- "learning_rate": 9.941557419012742e-06,
261
- "loss": 2.5382,
262
- "step": 360
263
- },
264
- {
265
- "epoch": 0.07945455521554733,
266
- "grad_norm": 13.076924324035645,
267
- "learning_rate": 9.936135303775702e-06,
268
- "loss": 1.6862,
269
- "step": 370
270
- },
271
- {
272
- "epoch": 0.08160197562677833,
273
- "grad_norm": 20.497676849365234,
274
- "learning_rate": 9.930474309972813e-06,
275
- "loss": 3.52,
276
- "step": 380
277
- },
278
- {
279
- "epoch": 0.08374939603800934,
280
- "grad_norm": 14.558286666870117,
281
- "learning_rate": 9.92457471156133e-06,
282
- "loss": 2.3912,
283
- "step": 390
284
- },
285
- {
286
- "epoch": 0.08589681644924035,
287
- "grad_norm": 12.070184707641602,
288
- "learning_rate": 9.918436794045507e-06,
289
- "loss": 1.997,
290
- "step": 400
291
- },
292
- {
293
- "epoch": 0.08804423686047136,
294
- "grad_norm": 13.852490425109863,
295
- "learning_rate": 9.912060854462766e-06,
296
- "loss": 3.0433,
297
- "step": 410
298
- },
299
- {
300
- "epoch": 0.09019165727170236,
301
- "grad_norm": 26.084009170532227,
302
- "learning_rate": 9.90544720136934e-06,
303
- "loss": 2.4177,
304
- "step": 420
305
- },
306
- {
307
- "epoch": 0.09233907768293338,
308
- "grad_norm": 12.258209228515625,
309
- "learning_rate": 9.898596154825333e-06,
310
- "loss": 1.9946,
311
- "step": 430
312
- },
313
- {
314
- "epoch": 0.09448649809416439,
315
- "grad_norm": 12.761174201965332,
316
- "learning_rate": 9.891508046379225e-06,
317
- "loss": 2.641,
318
- "step": 440
319
- },
320
- {
321
- "epoch": 0.09663391850539539,
322
- "grad_norm": 14.85302448272705,
323
- "learning_rate": 9.884183219051837e-06,
324
- "loss": 2.7384,
325
- "step": 450
326
- },
327
- {
328
- "epoch": 0.0987813389166264,
329
- "grad_norm": 13.194392204284668,
330
- "learning_rate": 9.876622027319726e-06,
331
- "loss": 1.9535,
332
- "step": 460
333
- },
334
- {
335
- "epoch": 0.10092875932785741,
336
- "grad_norm": 15.4808349609375,
337
- "learning_rate": 9.868824837098032e-06,
338
- "loss": 1.5539,
339
- "step": 470
340
- },
341
- {
342
- "epoch": 0.10307617973908842,
343
- "grad_norm": 14.890803337097168,
344
- "learning_rate": 9.860792025722768e-06,
345
- "loss": 2.2916,
346
- "step": 480
347
- },
348
- {
349
- "epoch": 0.10522360015031942,
350
- "grad_norm": 12.184898376464844,
351
- "learning_rate": 9.852523981932558e-06,
352
- "loss": 1.9207,
353
- "step": 490
354
- },
355
- {
356
- "epoch": 0.10737102056155044,
357
- "grad_norm": 10.492812156677246,
358
- "learning_rate": 9.844021105849837e-06,
359
- "loss": 1.8793,
360
- "step": 500
361
- },
362
- {
363
- "epoch": 0.10951844097278145,
364
- "grad_norm": 15.009350776672363,
365
- "learning_rate": 9.835283808961463e-06,
366
- "loss": 2.2671,
367
- "step": 510
368
- },
369
- {
370
- "epoch": 0.11166586138401245,
371
- "grad_norm": 11.420863151550293,
372
- "learning_rate": 9.82631251409883e-06,
373
- "loss": 1.9357,
374
- "step": 520
375
- },
376
- {
377
- "epoch": 0.11381328179524347,
378
- "grad_norm": 23.468830108642578,
379
- "learning_rate": 9.81710765541739e-06,
380
- "loss": 2.3423,
381
- "step": 530
382
- },
383
- {
384
- "epoch": 0.11596070220647448,
385
- "grad_norm": 11.901519775390625,
386
- "learning_rate": 9.807669678375643e-06,
387
- "loss": 2.2662,
388
- "step": 540
389
- },
390
- {
391
- "epoch": 0.11810812261770548,
392
- "grad_norm": 22.7636775970459,
393
- "learning_rate": 9.797999039713586e-06,
394
- "loss": 2.6903,
395
- "step": 550
396
- },
397
- {
398
- "epoch": 0.12025554302893648,
399
- "grad_norm": 10.768227577209473,
400
- "learning_rate": 9.788096207430608e-06,
401
- "loss": 1.8689,
402
- "step": 560
403
- },
404
- {
405
- "epoch": 0.1224029634401675,
406
- "grad_norm": 10.858819007873535,
407
- "learning_rate": 9.777961660762833e-06,
408
- "loss": 2.2812,
409
- "step": 570
410
- },
411
- {
412
- "epoch": 0.12455038385139851,
413
- "grad_norm": 21.28119468688965,
414
- "learning_rate": 9.767595890159944e-06,
415
- "loss": 2.6011,
416
- "step": 580
417
- },
418
- {
419
- "epoch": 0.12669780426262953,
420
- "grad_norm": 11.53591251373291,
421
- "learning_rate": 9.756999397261433e-06,
422
- "loss": 2.2605,
423
- "step": 590
424
- },
425
- {
426
- "epoch": 0.12884522467386053,
427
- "grad_norm": 9.207640647888184,
428
- "learning_rate": 9.746172694872332e-06,
429
- "loss": 1.4534,
430
- "step": 600
431
- },
432
- {
433
- "epoch": 0.13099264508509154,
434
- "grad_norm": 12.190791130065918,
435
- "learning_rate": 9.735116306938397e-06,
436
- "loss": 2.1114,
437
- "step": 610
438
- },
439
- {
440
- "epoch": 0.13314006549632254,
441
- "grad_norm": 11.150038719177246,
442
- "learning_rate": 9.72383076852075e-06,
443
- "loss": 2.1683,
444
- "step": 620
445
- },
446
- {
447
- "epoch": 0.13528748590755355,
448
- "grad_norm": 21.529489517211914,
449
- "learning_rate": 9.712316625769988e-06,
450
- "loss": 3.3042,
451
- "step": 630
452
- },
453
- {
454
- "epoch": 0.13743490631878455,
455
- "grad_norm": 11.192593574523926,
456
- "learning_rate": 9.700574435899745e-06,
457
- "loss": 1.8376,
458
- "step": 640
459
- },
460
- {
461
- "epoch": 0.13958232673001558,
462
- "grad_norm": 12.935983657836914,
463
- "learning_rate": 9.688604767159736e-06,
464
- "loss": 2.1633,
465
- "step": 650
466
- },
467
- {
468
- "epoch": 0.1417297471412466,
469
- "grad_norm": 11.339750289916992,
470
- "learning_rate": 9.676408198808253e-06,
471
- "loss": 3.0413,
472
- "step": 660
473
- },
474
- {
475
- "epoch": 0.1438771675524776,
476
- "grad_norm": 10.103887557983398,
477
- "learning_rate": 9.663985321084132e-06,
478
- "loss": 2.6125,
479
- "step": 670
480
- },
481
- {
482
- "epoch": 0.1460245879637086,
483
- "grad_norm": 20.008167266845703,
484
- "learning_rate": 9.651336735178191e-06,
485
- "loss": 1.8728,
486
- "step": 680
487
- },
488
- {
489
- "epoch": 0.1481720083749396,
490
- "grad_norm": 21.304391860961914,
491
- "learning_rate": 9.63846305320413e-06,
492
- "loss": 2.6488,
493
- "step": 690
494
- },
495
- {
496
- "epoch": 0.1503194287861706,
497
- "grad_norm": 11.066816329956055,
498
- "learning_rate": 9.62536489816892e-06,
499
- "loss": 2.2746,
500
- "step": 700
501
- },
502
- {
503
- "epoch": 0.1524668491974016,
504
- "grad_norm": 13.05954647064209,
505
- "learning_rate": 9.61204290394264e-06,
506
- "loss": 1.9458,
507
- "step": 710
508
- },
509
- {
510
- "epoch": 0.15461426960863264,
511
- "grad_norm": 10.509926795959473,
512
- "learning_rate": 9.598497715227815e-06,
513
- "loss": 1.7751,
514
- "step": 720
515
- },
516
- {
517
- "epoch": 0.15676169001986365,
518
- "grad_norm": 9.94949722290039,
519
- "learning_rate": 9.584729987528202e-06,
520
- "loss": 2.5051,
521
- "step": 730
522
- },
523
- {
524
- "epoch": 0.15890911043109465,
525
- "grad_norm": 10.857903480529785,
526
- "learning_rate": 9.570740387117078e-06,
527
- "loss": 2.6639,
528
- "step": 740
529
- },
530
- {
531
- "epoch": 0.16105653084232566,
532
- "grad_norm": 9.37006664276123,
533
- "learning_rate": 9.556529591005001e-06,
534
- "loss": 2.3239,
535
- "step": 750
536
- },
537
- {
538
- "epoch": 0.16320395125355666,
539
- "grad_norm": 12.794018745422363,
540
- "learning_rate": 9.542098286907024e-06,
541
- "loss": 1.7987,
542
- "step": 760
543
- },
544
- {
545
- "epoch": 0.16535137166478767,
546
- "grad_norm": 10.971644401550293,
547
- "learning_rate": 9.527447173209444e-06,
548
- "loss": 2.1475,
549
- "step": 770
550
- },
551
- {
552
- "epoch": 0.16749879207601867,
553
- "grad_norm": 12.863593101501465,
554
- "learning_rate": 9.51257695893598e-06,
555
- "loss": 2.5493,
556
- "step": 780
557
- },
558
- {
559
- "epoch": 0.1696462124872497,
560
- "grad_norm": 10.18968391418457,
561
- "learning_rate": 9.497488363713477e-06,
562
- "loss": 2.5998,
563
- "step": 790
564
- },
565
- {
566
- "epoch": 0.1717936328984807,
567
- "grad_norm": 10.925987243652344,
568
- "learning_rate": 9.482182117737066e-06,
569
- "loss": 1.7935,
570
- "step": 800
571
- },
572
- {
573
- "epoch": 0.1739410533097117,
574
- "grad_norm": 8.317191123962402,
575
- "learning_rate": 9.466658961734844e-06,
576
- "loss": 1.8083,
577
- "step": 810
578
- },
579
- {
580
- "epoch": 0.17608847372094272,
581
- "grad_norm": 11.145355224609375,
582
- "learning_rate": 9.450919646932013e-06,
583
- "loss": 2.2068,
584
- "step": 820
585
- },
586
- {
587
- "epoch": 0.17823589413217372,
588
- "grad_norm": 9.376663208007812,
589
- "learning_rate": 9.434964935014527e-06,
590
- "loss": 2.5394,
591
- "step": 830
592
- },
593
- {
594
- "epoch": 0.18038331454340473,
595
- "grad_norm": 10.962883949279785,
596
- "learning_rate": 9.418795598092243e-06,
597
- "loss": 2.5965,
598
- "step": 840
599
- },
600
- {
601
- "epoch": 0.18253073495463573,
602
- "grad_norm": 9.319477081298828,
603
- "learning_rate": 9.402412418661541e-06,
604
- "loss": 2.9125,
605
- "step": 850
606
- },
607
- {
608
- "epoch": 0.18467815536586676,
609
- "grad_norm": 10.956426620483398,
610
- "learning_rate": 9.385816189567462e-06,
611
- "loss": 2.168,
612
- "step": 860
613
- },
614
- {
615
- "epoch": 0.18682557577709777,
616
- "grad_norm": 11.152973175048828,
617
- "learning_rate": 9.369007713965346e-06,
618
- "loss": 1.8573,
619
- "step": 870
620
- },
621
- {
622
- "epoch": 0.18897299618832877,
623
- "grad_norm": 10.0717134475708,
624
- "learning_rate": 9.351987805281949e-06,
625
- "loss": 2.5868,
626
- "step": 880
627
- },
628
- {
629
- "epoch": 0.19112041659955978,
630
- "grad_norm": 15.084976196289062,
631
- "learning_rate": 9.334757287176089e-06,
632
- "loss": 2.2103,
633
- "step": 890
634
- },
635
- {
636
- "epoch": 0.19326783701079078,
637
- "grad_norm": 10.498251914978027,
638
- "learning_rate": 9.317316993498788e-06,
639
- "loss": 1.8223,
640
- "step": 900
641
- },
642
- {
643
- "epoch": 0.1954152574220218,
644
- "grad_norm": 9.38797664642334,
645
- "learning_rate": 9.299667768252911e-06,
646
- "loss": 1.7798,
647
- "step": 910
648
- },
649
- {
650
- "epoch": 0.1975626778332528,
651
- "grad_norm": 10.024442672729492,
652
- "learning_rate": 9.281810465552327e-06,
653
- "loss": 1.4073,
654
- "step": 920
655
- },
656
- {
657
- "epoch": 0.19971009824448382,
658
- "grad_norm": 9.467859268188477,
659
- "learning_rate": 9.263745949580569e-06,
660
- "loss": 2.2005,
661
- "step": 930
662
- },
663
- {
664
- "epoch": 0.20185751865571483,
665
- "grad_norm": 10.421006202697754,
666
- "learning_rate": 9.24547509454902e-06,
667
- "loss": 1.4132,
668
- "step": 940
669
- },
670
- {
671
- "epoch": 0.20400493906694583,
672
- "grad_norm": 10.632217407226562,
673
- "learning_rate": 9.226998784654606e-06,
674
- "loss": 2.5512,
675
- "step": 950
676
- },
677
- {
678
- "epoch": 0.20615235947817684,
679
- "grad_norm": 10.489507675170898,
680
- "learning_rate": 9.208317914036997e-06,
681
- "loss": 1.8359,
682
- "step": 960
683
- },
684
- {
685
- "epoch": 0.20829977988940784,
686
- "grad_norm": 22.003978729248047,
687
- "learning_rate": 9.18943338673535e-06,
688
- "loss": 2.1296,
689
- "step": 970
690
- },
691
- {
692
- "epoch": 0.21044720030063885,
693
- "grad_norm": 9.565354347229004,
694
- "learning_rate": 9.170346116644545e-06,
695
- "loss": 2.1198,
696
- "step": 980
697
- },
698
- {
699
- "epoch": 0.21259462071186985,
700
- "grad_norm": 10.454195022583008,
701
- "learning_rate": 9.15105702747097e-06,
702
- "loss": 2.2026,
703
- "step": 990
704
- },
705
- {
706
- "epoch": 0.21474204112310089,
707
- "grad_norm": 19.8426456451416,
708
- "learning_rate": 9.131567052687811e-06,
709
- "loss": 2.9161,
710
- "step": 1000
711
- },
712
- {
713
- "epoch": 0.2168894615343319,
714
- "grad_norm": 19.293813705444336,
715
- "learning_rate": 9.111877135489886e-06,
716
- "loss": 2.5228,
717
- "step": 1010
718
- },
719
- {
720
- "epoch": 0.2190368819455629,
721
- "grad_norm": 10.029641151428223,
722
- "learning_rate": 9.091988228747992e-06,
723
- "loss": 1.869,
724
- "step": 1020
725
- },
726
- {
727
- "epoch": 0.2211843023567939,
728
- "grad_norm": 9.284717559814453,
729
- "learning_rate": 9.071901294962795e-06,
730
- "loss": 3.7155,
731
- "step": 1030
732
- },
733
- {
734
- "epoch": 0.2233317227680249,
735
- "grad_norm": 19.013784408569336,
736
- "learning_rate": 9.05161730621825e-06,
737
- "loss": 2.1832,
738
- "step": 1040
739
- },
740
- {
741
- "epoch": 0.2254791431792559,
742
- "grad_norm": 9.73091983795166,
743
- "learning_rate": 9.03113724413456e-06,
744
- "loss": 2.1185,
745
- "step": 1050
746
- },
747
- {
748
- "epoch": 0.22762656359048694,
749
- "grad_norm": 10.244536399841309,
750
- "learning_rate": 9.010462099820674e-06,
751
- "loss": 1.7729,
752
- "step": 1060
753
- },
754
- {
755
- "epoch": 0.22977398400171795,
756
- "grad_norm": 8.954618453979492,
757
- "learning_rate": 8.989592873826316e-06,
758
- "loss": 1.7244,
759
- "step": 1070
760
- },
761
- {
762
- "epoch": 0.23192140441294895,
763
- "grad_norm": 19.035173416137695,
764
- "learning_rate": 8.96853057609357e-06,
765
- "loss": 2.5808,
766
- "step": 1080
767
- },
768
- {
769
- "epoch": 0.23406882482417996,
770
- "grad_norm": 9.041604042053223,
771
- "learning_rate": 8.947276225908001e-06,
772
- "loss": 1.7619,
773
- "step": 1090
774
- },
775
- {
776
- "epoch": 0.23621624523541096,
777
- "grad_norm": 11.142961502075195,
778
- "learning_rate": 8.925830851849338e-06,
779
- "loss": 1.7543,
780
- "step": 1100
781
- },
782
- {
783
- "epoch": 0.23836366564664196,
784
- "grad_norm": 8.855928421020508,
785
- "learning_rate": 8.904195491741682e-06,
786
- "loss": 1.4371,
787
- "step": 1110
788
- },
789
- {
790
- "epoch": 0.24051108605787297,
791
- "grad_norm": 10.576139450073242,
792
- "learning_rate": 8.882371192603297e-06,
793
- "loss": 1.7964,
794
- "step": 1120
795
- },
796
- {
797
- "epoch": 0.242658506469104,
798
- "grad_norm": 19.504796981811523,
799
- "learning_rate": 8.86035901059592e-06,
800
- "loss": 2.4225,
801
- "step": 1130
802
- },
803
- {
804
- "epoch": 0.244805926880335,
805
- "grad_norm": 8.53579330444336,
806
- "learning_rate": 8.83816001097368e-06,
807
- "loss": 2.5166,
808
- "step": 1140
809
- },
810
- {
811
- "epoch": 0.246953347291566,
812
- "grad_norm": 10.541763305664062,
813
- "learning_rate": 8.815775268031514e-06,
814
- "loss": 2.5292,
815
- "step": 1150
816
- },
817
- {
818
- "epoch": 0.24910076770279702,
819
- "grad_norm": 8.700005531311035,
820
- "learning_rate": 8.7932058650532e-06,
821
- "loss": 1.7864,
822
- "step": 1160
823
- },
824
- {
825
- "epoch": 0.25124818811402805,
826
- "grad_norm": 19.96916961669922,
827
- "learning_rate": 8.770452894258922e-06,
828
- "loss": 2.1556,
829
- "step": 1170
830
- },
831
- {
832
- "epoch": 0.25339560852525905,
833
- "grad_norm": 10.887153625488281,
834
- "learning_rate": 8.747517456752419e-06,
835
- "loss": 1.7778,
836
- "step": 1180
837
- },
838
- {
839
- "epoch": 0.25554302893649006,
840
- "grad_norm": 9.90004825592041,
841
- "learning_rate": 8.724400662467694e-06,
842
- "loss": 2.4839,
843
- "step": 1190
844
- },
845
- {
846
- "epoch": 0.25769044934772106,
847
- "grad_norm": 18.490222930908203,
848
- "learning_rate": 8.701103630115303e-06,
849
- "loss": 2.9508,
850
- "step": 1200
851
- },
852
- {
853
- "epoch": 0.25983786975895207,
854
- "grad_norm": 10.261496543884277,
855
- "learning_rate": 8.677627487128211e-06,
856
- "loss": 2.1185,
857
- "step": 1210
858
- },
859
- {
860
- "epoch": 0.26198529017018307,
861
- "grad_norm": 11.693336486816406,
862
- "learning_rate": 8.65397336960724e-06,
863
- "loss": 2.5583,
864
- "step": 1220
865
- },
866
- {
867
- "epoch": 0.2641327105814141,
868
- "grad_norm": 9.892570495605469,
869
- "learning_rate": 8.630142422266086e-06,
870
- "loss": 2.5668,
871
- "step": 1230
872
- },
873
- {
874
- "epoch": 0.2662801309926451,
875
- "grad_norm": 11.391189575195312,
876
- "learning_rate": 8.60613579837591e-06,
877
- "loss": 2.1724,
878
- "step": 1240
879
- },
880
- {
881
- "epoch": 0.2684275514038761,
882
- "grad_norm": 9.438261032104492,
883
- "learning_rate": 8.581954659709549e-06,
884
- "loss": 2.51,
885
- "step": 1250
886
- },
887
- {
888
- "epoch": 0.2705749718151071,
889
- "grad_norm": 12.741863250732422,
890
- "learning_rate": 8.55760017648527e-06,
891
- "loss": 2.1885,
892
- "step": 1260
893
- },
894
- {
895
- "epoch": 0.2727223922263381,
896
- "grad_norm": 21.02118492126465,
897
- "learning_rate": 8.533073527310157e-06,
898
- "loss": 2.5175,
899
- "step": 1270
900
- },
901
- {
902
- "epoch": 0.2748698126375691,
903
- "grad_norm": 10.651395797729492,
904
- "learning_rate": 8.508375899123064e-06,
905
- "loss": 2.1997,
906
- "step": 1280
907
- },
908
- {
909
- "epoch": 0.2770172330488001,
910
- "grad_norm": 10.738212585449219,
911
- "learning_rate": 8.483508487137175e-06,
912
- "loss": 2.1453,
913
- "step": 1290
914
- },
915
- {
916
- "epoch": 0.27916465346003116,
917
- "grad_norm": 14.632546424865723,
918
- "learning_rate": 8.458472494782169e-06,
919
- "loss": 2.1023,
920
- "step": 1300
921
- },
922
- {
923
- "epoch": 0.28131207387126217,
924
- "grad_norm": 8.816224098205566,
925
- "learning_rate": 8.433269133645974e-06,
926
- "loss": 1.682,
927
- "step": 1310
928
- },
929
- {
930
- "epoch": 0.2834594942824932,
931
- "grad_norm": 8.408437728881836,
932
- "learning_rate": 8.407899623416136e-06,
933
- "loss": 1.7611,
934
- "step": 1320
935
- },
936
- {
937
- "epoch": 0.2856069146937242,
938
- "grad_norm": 9.93062686920166,
939
- "learning_rate": 8.382365191820796e-06,
940
- "loss": 2.503,
941
- "step": 1330
942
- },
943
- {
944
- "epoch": 0.2877543351049552,
945
- "grad_norm": 11.0846586227417,
946
- "learning_rate": 8.356667074569274e-06,
947
- "loss": 2.0817,
948
- "step": 1340
949
- },
950
- {
951
- "epoch": 0.2899017555161862,
952
- "grad_norm": 18.90940284729004,
953
- "learning_rate": 8.330806515292271e-06,
954
- "loss": 2.1409,
955
- "step": 1350
956
- },
957
- {
958
- "epoch": 0.2920491759274172,
959
- "grad_norm": 7.251021862030029,
960
- "learning_rate": 8.304784765481676e-06,
961
- "loss": 1.7067,
962
- "step": 1360
963
- },
964
- {
965
- "epoch": 0.2941965963386482,
966
- "grad_norm": 18.02667808532715,
967
- "learning_rate": 8.278603084430012e-06,
968
- "loss": 2.5272,
969
- "step": 1370
970
- },
971
- {
972
- "epoch": 0.2963440167498792,
973
- "grad_norm": 19.2054386138916,
974
- "learning_rate": 8.25226273916949e-06,
975
- "loss": 2.1962,
976
- "step": 1380
977
- },
978
- {
979
- "epoch": 0.2984914371611102,
980
- "grad_norm": 8.991003036499023,
981
- "learning_rate": 8.225765004410688e-06,
982
- "loss": 2.8256,
983
- "step": 1390
984
- },
985
- {
986
- "epoch": 0.3006388575723412,
987
- "grad_norm": 10.063870429992676,
988
- "learning_rate": 8.199111162480871e-06,
989
- "loss": 2.5387,
990
- "step": 1400
991
- },
992
- {
993
- "epoch": 0.3027862779835722,
994
- "grad_norm": 9.316102027893066,
995
- "learning_rate": 8.17230250326193e-06,
996
- "loss": 1.7767,
997
- "step": 1410
998
- },
999
- {
1000
- "epoch": 0.3049336983948032,
1001
- "grad_norm": 18.266010284423828,
1002
- "learning_rate": 8.145340324127958e-06,
1003
- "loss": 1.7364,
1004
- "step": 1420
1005
- },
1006
- {
1007
- "epoch": 0.3070811188060342,
1008
- "grad_norm": 18.232187271118164,
1009
- "learning_rate": 8.118225929882468e-06,
1010
- "loss": 2.1927,
1011
- "step": 1430
1012
- },
1013
- {
1014
- "epoch": 0.3092285392172653,
1015
- "grad_norm": 8.176814079284668,
1016
- "learning_rate": 8.090960632695246e-06,
1017
- "loss": 2.1411,
1018
- "step": 1440
1019
- },
1020
- {
1021
- "epoch": 0.3113759596284963,
1022
- "grad_norm": 9.711817741394043,
1023
- "learning_rate": 8.063545752038854e-06,
1024
- "loss": 1.7321,
1025
- "step": 1450
1026
- },
1027
- {
1028
- "epoch": 0.3135233800397273,
1029
- "grad_norm": 11.406618118286133,
1030
- "learning_rate": 8.035982614624774e-06,
1031
- "loss": 1.3379,
1032
- "step": 1460
1033
- },
1034
- {
1035
- "epoch": 0.3156708004509583,
1036
- "grad_norm": 8.103960990905762,
1037
- "learning_rate": 8.008272554339195e-06,
1038
- "loss": 2.4391,
1039
- "step": 1470
1040
- },
1041
- {
1042
- "epoch": 0.3178182208621893,
1043
- "grad_norm": 10.89107608795166,
1044
- "learning_rate": 7.980416912178478e-06,
1045
- "loss": 2.6215,
1046
- "step": 1480
1047
- },
1048
- {
1049
- "epoch": 0.3199656412734203,
1050
- "grad_norm": 13.625832557678223,
1051
- "learning_rate": 7.952417036184248e-06,
1052
- "loss": 1.3208,
1053
- "step": 1490
1054
- },
1055
- {
1056
- "epoch": 0.3221130616846513,
1057
- "grad_norm": 9.747784614562988,
1058
- "learning_rate": 7.924274281378153e-06,
1059
- "loss": 1.3278,
1060
- "step": 1500
1061
- },
1062
- {
1063
- "epoch": 0.3242604820958823,
1064
- "grad_norm": 23.162227630615234,
1065
- "learning_rate": 7.895990009696308e-06,
1066
- "loss": 2.5893,
1067
- "step": 1510
1068
- },
1069
- {
1070
- "epoch": 0.3264079025071133,
1071
- "grad_norm": 9.394268989562988,
1072
- "learning_rate": 7.867565589923364e-06,
1073
- "loss": 2.1448,
1074
- "step": 1520
1075
- },
1076
- {
1077
- "epoch": 0.32855532291834433,
1078
- "grad_norm": 8.258391380310059,
1079
- "learning_rate": 7.839002397626281e-06,
1080
- "loss": 2.1836,
1081
- "step": 1530
1082
- },
1083
- {
1084
- "epoch": 0.33070274332957533,
1085
- "grad_norm": 10.303350448608398,
1086
- "learning_rate": 7.810301815087753e-06,
1087
- "loss": 3.2407,
1088
- "step": 1540
1089
- },
1090
- {
1091
- "epoch": 0.33285016374080634,
1092
- "grad_norm": 8.373433113098145,
1093
- "learning_rate": 7.781465231239318e-06,
1094
- "loss": 2.5282,
1095
- "step": 1550
1096
- },
1097
- {
1098
- "epoch": 0.33499758415203734,
1099
- "grad_norm": 8.955942153930664,
1100
- "learning_rate": 7.75249404159414e-06,
1101
- "loss": 2.4664,
1102
- "step": 1560
1103
- },
1104
- {
1105
- "epoch": 0.33714500456326835,
1106
- "grad_norm": 8.981405258178711,
1107
- "learning_rate": 7.723389648179475e-06,
1108
- "loss": 2.0393,
1109
- "step": 1570
1110
- },
1111
- {
1112
- "epoch": 0.3392924249744994,
1113
- "grad_norm": 11.585493087768555,
1114
- "learning_rate": 7.694153459468822e-06,
1115
- "loss": 1.7473,
1116
- "step": 1580
1117
- },
1118
- {
1119
- "epoch": 0.3414398453857304,
1120
- "grad_norm": 7.84296989440918,
1121
- "learning_rate": 7.664786890313757e-06,
1122
- "loss": 2.0628,
1123
- "step": 1590
1124
- },
1125
- {
1126
- "epoch": 0.3435872657969614,
1127
- "grad_norm": 7.5332417488098145,
1128
- "learning_rate": 7.635291361875474e-06,
1129
- "loss": 2.1288,
1130
- "step": 1600
1131
- },
1132
- {
1133
- "epoch": 0.3457346862081924,
1134
- "grad_norm": 10.541316986083984,
1135
- "learning_rate": 7.6056683015559975e-06,
1136
- "loss": 1.7313,
1137
- "step": 1610
1138
- },
1139
- {
1140
- "epoch": 0.3478821066194234,
1141
- "grad_norm": 8.132133483886719,
1142
- "learning_rate": 7.57591914292911e-06,
1143
- "loss": 1.7268,
1144
- "step": 1620
1145
- },
1146
- {
1147
- "epoch": 0.35002952703065443,
1148
- "grad_norm": 20.494726181030273,
1149
- "learning_rate": 7.546045325670979e-06,
1150
- "loss": 3.3231,
1151
- "step": 1630
1152
- },
1153
- {
1154
- "epoch": 0.35217694744188544,
1155
- "grad_norm": 10.624666213989258,
1156
- "learning_rate": 7.516048295490479e-06,
1157
- "loss": 2.0748,
1158
- "step": 1640
1159
- },
1160
- {
1161
- "epoch": 0.35432436785311644,
1162
- "grad_norm": 9.553028106689453,
1163
- "learning_rate": 7.485929504059234e-06,
1164
- "loss": 1.6695,
1165
- "step": 1650
1166
- },
1167
- {
1168
- "epoch": 0.35647178826434744,
1169
- "grad_norm": 8.269372940063477,
1170
- "learning_rate": 7.455690408941363e-06,
1171
- "loss": 2.1454,
1172
- "step": 1660
1173
- },
1174
- {
1175
- "epoch": 0.35861920867557845,
1176
- "grad_norm": 8.983115196228027,
1177
- "learning_rate": 7.425332473522942e-06,
1178
- "loss": 2.085,
1179
- "step": 1670
1180
- },
1181
- {
1182
- "epoch": 0.36076662908680945,
1183
- "grad_norm": 10.12364387512207,
1184
- "learning_rate": 7.394857166941187e-06,
1185
- "loss": 1.7961,
1186
- "step": 1680
1187
- },
1188
- {
1189
- "epoch": 0.36291404949804046,
1190
- "grad_norm": 13.731148719787598,
1191
- "learning_rate": 7.364265964013355e-06,
1192
- "loss": 1.8293,
1193
- "step": 1690
1194
- },
1195
- {
1196
- "epoch": 0.36506146990927146,
1197
- "grad_norm": 7.890660762786865,
1198
- "learning_rate": 7.333560345165371e-06,
1199
- "loss": 1.694,
1200
- "step": 1700
1201
- },
1202
- {
1203
- "epoch": 0.3672088903205025,
1204
- "grad_norm": 7.874722957611084,
1205
- "learning_rate": 7.302741796360192e-06,
1206
- "loss": 2.1216,
1207
- "step": 1710
1208
- },
1209
- {
1210
- "epoch": 0.36935631073173353,
1211
- "grad_norm": 9.214128494262695,
1212
- "learning_rate": 7.271811809025882e-06,
1213
- "loss": 1.3505,
1214
- "step": 1720
1215
- },
1216
- {
1217
- "epoch": 0.37150373114296453,
1218
- "grad_norm": 8.188569068908691,
1219
- "learning_rate": 7.240771879983451e-06,
1220
- "loss": 2.1366,
1221
- "step": 1730
1222
- },
1223
- {
1224
- "epoch": 0.37365115155419554,
1225
- "grad_norm": 7.846069812774658,
1226
- "learning_rate": 7.209623511374407e-06,
1227
- "loss": 1.351,
1228
- "step": 1740
1229
- },
1230
- {
1231
- "epoch": 0.37579857196542654,
1232
- "grad_norm": 11.016289710998535,
1233
- "learning_rate": 7.178368210588067e-06,
1234
- "loss": 2.1055,
1235
- "step": 1750
1236
- },
1237
- {
1238
- "epoch": 0.37794599237665755,
1239
- "grad_norm": 9.798548698425293,
1240
- "learning_rate": 7.1470074901886065e-06,
1241
- "loss": 1.6451,
1242
- "step": 1760
1243
- },
1244
- {
1245
- "epoch": 0.38009341278788855,
1246
- "grad_norm": 19.033615112304688,
1247
- "learning_rate": 7.1155428678418615e-06,
1248
- "loss": 2.1027,
1249
- "step": 1770
1250
- },
1251
- {
1252
- "epoch": 0.38224083319911956,
1253
- "grad_norm": 10.220734596252441,
1254
- "learning_rate": 7.083975866241881e-06,
1255
- "loss": 2.43,
1256
- "step": 1780
1257
- },
1258
- {
1259
- "epoch": 0.38438825361035056,
1260
- "grad_norm": 17.87563705444336,
1261
- "learning_rate": 7.0523080130372444e-06,
1262
- "loss": 3.0401,
1263
- "step": 1790
1264
- },
1265
- {
1266
- "epoch": 0.38653567402158157,
1267
- "grad_norm": 7.5210161209106445,
1268
- "learning_rate": 7.020540840757124e-06,
1269
- "loss": 2.4589,
1270
- "step": 1800
1271
- },
1272
- {
1273
- "epoch": 0.38868309443281257,
1274
- "grad_norm": 8.672847747802734,
1275
- "learning_rate": 6.988675886737125e-06,
1276
- "loss": 1.6752,
1277
- "step": 1810
1278
- },
1279
- {
1280
- "epoch": 0.3908305148440436,
1281
- "grad_norm": 8.692748069763184,
1282
- "learning_rate": 6.956714693044888e-06,
1283
- "loss": 1.7287,
1284
- "step": 1820
1285
- },
1286
- {
1287
- "epoch": 0.3929779352552746,
1288
- "grad_norm": 7.074733257293701,
1289
- "learning_rate": 6.9246588064054575e-06,
1290
- "loss": 2.8803,
1291
- "step": 1830
1292
- },
1293
- {
1294
- "epoch": 0.3951253556665056,
1295
- "grad_norm": 18.272645950317383,
1296
- "learning_rate": 6.892509778126442e-06,
1297
- "loss": 3.2607,
1298
- "step": 1840
1299
- },
1300
- {
1301
- "epoch": 0.39727277607773664,
1302
- "grad_norm": 9.085878372192383,
1303
- "learning_rate": 6.860269164022921e-06,
1304
- "loss": 1.7211,
1305
- "step": 1850
1306
- },
1307
- {
1308
- "epoch": 0.39942019648896765,
1309
- "grad_norm": 7.339599132537842,
1310
- "learning_rate": 6.827938524342175e-06,
1311
- "loss": 1.6545,
1312
- "step": 1860
1313
- },
1314
- {
1315
- "epoch": 0.40156761690019865,
1316
- "grad_norm": 18.716262817382812,
1317
- "learning_rate": 6.7955194236881595e-06,
1318
- "loss": 2.499,
1319
- "step": 1870
1320
- },
1321
- {
1322
- "epoch": 0.40371503731142966,
1323
- "grad_norm": 10.822924613952637,
1324
- "learning_rate": 6.763013430945803e-06,
1325
- "loss": 2.7362,
1326
- "step": 1880
1327
- },
1328
- {
1329
- "epoch": 0.40586245772266066,
1330
- "grad_norm": 8.549121856689453,
1331
- "learning_rate": 6.730422119205072e-06,
1332
- "loss": 2.4742,
1333
- "step": 1890
1334
- },
1335
- {
1336
- "epoch": 0.40800987813389167,
1337
- "grad_norm": 8.39169692993164,
1338
- "learning_rate": 6.697747065684851e-06,
1339
- "loss": 2.0802,
1340
- "step": 1900
1341
- },
1342
- {
1343
- "epoch": 0.4101572985451227,
1344
- "grad_norm": 10.248717308044434,
1345
- "learning_rate": 6.664989851656606e-06,
1346
- "loss": 2.1458,
1347
- "step": 1910
1348
- },
1349
- {
1350
- "epoch": 0.4123047189563537,
1351
- "grad_norm": 9.720662117004395,
1352
- "learning_rate": 6.632152062367871e-06,
1353
- "loss": 1.6682,
1354
- "step": 1920
1355
- },
1356
- {
1357
- "epoch": 0.4144521393675847,
1358
- "grad_norm": 9.940651893615723,
1359
- "learning_rate": 6.599235286965524e-06,
1360
- "loss": 1.6989,
1361
- "step": 1930
1362
- },
1363
- {
1364
- "epoch": 0.4165995597788157,
1365
- "grad_norm": 9.139376640319824,
1366
- "learning_rate": 6.566241118418888e-06,
1367
- "loss": 2.0926,
1368
- "step": 1940
1369
- },
1370
- {
1371
- "epoch": 0.4187469801900467,
1372
- "grad_norm": 8.232985496520996,
1373
- "learning_rate": 6.5331711534426326e-06,
1374
- "loss": 2.4193,
1375
- "step": 1950
1376
- },
1377
- {
1378
- "epoch": 0.4208944006012777,
1379
- "grad_norm": 7.869434356689453,
1380
- "learning_rate": 6.50002699241951e-06,
1381
- "loss": 2.5119,
1382
- "step": 1960
1383
- },
1384
- {
1385
- "epoch": 0.4230418210125087,
1386
- "grad_norm": 9.408770561218262,
1387
- "learning_rate": 6.466810239322909e-06,
1388
- "loss": 1.3976,
1389
- "step": 1970
1390
- },
1391
- {
1392
- "epoch": 0.4251892414237397,
1393
- "grad_norm": 8.584155082702637,
1394
- "learning_rate": 6.43352250163922e-06,
1395
- "loss": 1.6854,
1396
- "step": 1980
1397
- },
1398
- {
1399
- "epoch": 0.42733666183497077,
1400
- "grad_norm": 18.368183135986328,
1401
- "learning_rate": 6.4001653902900564e-06,
1402
- "loss": 2.9371,
1403
- "step": 1990
1404
- },
1405
- {
1406
- "epoch": 0.42948408224620177,
1407
- "grad_norm": 10.570369720458984,
1408
- "learning_rate": 6.366740519554286e-06,
1409
- "loss": 1.3262,
1410
- "step": 2000
1411
- },
1412
- {
1413
- "epoch": 0.4316315026574328,
1414
- "grad_norm": 10.918615341186523,
1415
- "learning_rate": 6.33324950698992e-06,
1416
- "loss": 2.1305,
1417
- "step": 2010
1418
- },
1419
- {
1420
- "epoch": 0.4337789230686638,
1421
- "grad_norm": 9.189411163330078,
1422
- "learning_rate": 6.299693973355821e-06,
1423
- "loss": 2.079,
1424
- "step": 2020
1425
- },
1426
- {
1427
- "epoch": 0.4359263434798948,
1428
- "grad_norm": 19.311662673950195,
1429
- "learning_rate": 6.266075542533275e-06,
1430
- "loss": 2.4221,
1431
- "step": 2030
1432
- },
1433
- {
1434
- "epoch": 0.4380737638911258,
1435
- "grad_norm": 10.08248233795166,
1436
- "learning_rate": 6.2323958414474065e-06,
1437
- "loss": 1.703,
1438
- "step": 2040
1439
- },
1440
- {
1441
- "epoch": 0.4402211843023568,
1442
- "grad_norm": 18.993967056274414,
1443
- "learning_rate": 6.198656499988444e-06,
1444
- "loss": 1.9942,
1445
- "step": 2050
1446
- },
1447
- {
1448
- "epoch": 0.4423686047135878,
1449
- "grad_norm": 18.06959342956543,
1450
- "learning_rate": 6.164859150932839e-06,
1451
- "loss": 2.7462,
1452
- "step": 2060
1453
- },
1454
- {
1455
- "epoch": 0.4445160251248188,
1456
- "grad_norm": 17.054033279418945,
1457
- "learning_rate": 6.131005429864262e-06,
1458
- "loss": 2.7662,
1459
- "step": 2070
1460
- },
1461
- {
1462
- "epoch": 0.4466634455360498,
1463
- "grad_norm": 22.749589920043945,
1464
- "learning_rate": 6.097096975094432e-06,
1465
- "loss": 2.0372,
1466
- "step": 2080
1467
- },
1468
- {
1469
- "epoch": 0.4488108659472808,
1470
- "grad_norm": 8.990077018737793,
1471
- "learning_rate": 6.063135427583849e-06,
1472
- "loss": 2.3963,
1473
- "step": 2090
1474
- },
1475
- {
1476
- "epoch": 0.4509582863585118,
1477
- "grad_norm": 8.372001647949219,
1478
- "learning_rate": 6.029122430862373e-06,
1479
- "loss": 1.7628,
1480
- "step": 2100
1481
- },
1482
- {
1483
- "epoch": 0.4531057067697428,
1484
- "grad_norm": 8.995088577270508,
1485
- "learning_rate": 5.9950596309496864e-06,
1486
- "loss": 1.6897,
1487
- "step": 2110
1488
- },
1489
- {
1490
- "epoch": 0.4552531271809739,
1491
- "grad_norm": 10.031561851501465,
1492
- "learning_rate": 5.9609486762756465e-06,
1493
- "loss": 1.7369,
1494
- "step": 2120
1495
- },
1496
- {
1497
- "epoch": 0.4574005475922049,
1498
- "grad_norm": 8.10063362121582,
1499
- "learning_rate": 5.9267912176004925e-06,
1500
- "loss": 2.051,
1501
- "step": 2130
1502
- },
1503
- {
1504
- "epoch": 0.4595479680034359,
1505
- "grad_norm": 9.510966300964355,
1506
- "learning_rate": 5.892588907934988e-06,
1507
- "loss": 2.4391,
1508
- "step": 2140
1509
- },
1510
- {
1511
- "epoch": 0.4616953884146669,
1512
- "grad_norm": 19.641340255737305,
1513
- "learning_rate": 5.858343402460391e-06,
1514
- "loss": 2.0852,
1515
- "step": 2150
1516
- },
1517
- {
1518
- "epoch": 0.4638428088258979,
1519
- "grad_norm": 8.368480682373047,
1520
- "learning_rate": 5.8240563584483855e-06,
1521
- "loss": 2.0298,
1522
- "step": 2160
1523
- },
1524
- {
1525
- "epoch": 0.4659902292371289,
1526
- "grad_norm": 18.213680267333984,
1527
- "learning_rate": 5.789729435180853e-06,
1528
- "loss": 2.4393,
1529
- "step": 2170
1530
- },
1531
- {
1532
- "epoch": 0.4681376496483599,
1533
- "grad_norm": 17.781763076782227,
1534
- "learning_rate": 5.7553642938695945e-06,
1535
- "loss": 1.6106,
1536
- "step": 2180
1537
- },
1538
- {
1539
- "epoch": 0.4702850700595909,
1540
- "grad_norm": 8.500686645507812,
1541
- "learning_rate": 5.720962597575922e-06,
1542
- "loss": 1.6916,
1543
- "step": 2190
1544
- },
1545
- {
1546
- "epoch": 0.4724324904708219,
1547
- "grad_norm": 8.981590270996094,
1548
- "learning_rate": 5.68652601113019e-06,
1549
- "loss": 1.2966,
1550
- "step": 2200
1551
- },
1552
- {
1553
- "epoch": 0.4745799108820529,
1554
- "grad_norm": 17.556732177734375,
1555
- "learning_rate": 5.652056201051217e-06,
1556
- "loss": 2.4725,
1557
- "step": 2210
1558
- },
1559
- {
1560
- "epoch": 0.47672733129328393,
1561
- "grad_norm": 8.62265396118164,
1562
- "learning_rate": 5.617554835465646e-06,
1563
- "loss": 1.6889,
1564
- "step": 2220
1565
- },
1566
- {
1567
- "epoch": 0.47887475170451493,
1568
- "grad_norm": 19.104825973510742,
1569
- "learning_rate": 5.583023584027204e-06,
1570
- "loss": 2.4591,
1571
- "step": 2230
1572
- },
1573
- {
1574
- "epoch": 0.48102217211574594,
1575
- "grad_norm": 8.052384376525879,
1576
- "learning_rate": 5.548464117835917e-06,
1577
- "loss": 2.127,
1578
- "step": 2240
1579
- },
1580
- {
1581
- "epoch": 0.48316959252697694,
1582
- "grad_norm": 17.943681716918945,
1583
- "learning_rate": 5.513878109357228e-06,
1584
- "loss": 2.1374,
1585
- "step": 2250
1586
- },
1587
- {
1588
- "epoch": 0.485317012938208,
1589
- "grad_norm": 8.557195663452148,
1590
- "learning_rate": 5.479267232341064e-06,
1591
- "loss": 2.1146,
1592
- "step": 2260
1593
- },
1594
- {
1595
- "epoch": 0.487464433349439,
1596
- "grad_norm": 8.944927215576172,
1597
- "learning_rate": 5.444633161740834e-06,
1598
- "loss": 3.1203,
1599
- "step": 2270
1600
- },
1601
- {
1602
- "epoch": 0.48961185376067,
1603
- "grad_norm": 16.947784423828125,
1604
- "learning_rate": 5.40997757363238e-06,
1605
- "loss": 2.7593,
1606
- "step": 2280
1607
- },
1608
- {
1609
- "epoch": 0.491759274171901,
1610
- "grad_norm": 7.79644250869751,
1611
- "learning_rate": 5.3753021451328525e-06,
1612
- "loss": 2.0236,
1613
- "step": 2290
1614
- },
1615
- {
1616
- "epoch": 0.493906694583132,
1617
- "grad_norm": 7.248489856719971,
1618
- "learning_rate": 5.3406085543195555e-06,
1619
- "loss": 1.6521,
1620
- "step": 2300
1621
- },
1622
- {
1623
- "epoch": 0.496054114994363,
1624
- "grad_norm": 17.251184463500977,
1625
- "learning_rate": 5.305898480148741e-06,
1626
- "loss": 1.9537,
1627
- "step": 2310
1628
- },
1629
- {
1630
- "epoch": 0.49820153540559403,
1631
- "grad_norm": 8.320523262023926,
1632
- "learning_rate": 5.27117360237435e-06,
1633
- "loss": 2.8172,
1634
- "step": 2320
1635
- },
1636
- {
1637
- "epoch": 0.500348955816825,
1638
- "grad_norm": 7.883960723876953,
1639
- "learning_rate": 5.23643560146673e-06,
1640
- "loss": 2.7325,
1641
- "step": 2330
1642
- },
1643
- {
1644
- "epoch": 0.5024963762280561,
1645
- "grad_norm": 17.11644172668457,
1646
- "learning_rate": 5.201686158531304e-06,
1647
- "loss": 2.0326,
1648
- "step": 2340
1649
- },
1650
- {
1651
- "epoch": 0.504643796639287,
1652
- "grad_norm": 9.46158218383789,
1653
- "learning_rate": 5.166926955227224e-06,
1654
- "loss": 2.4668,
1655
- "step": 2350
1656
- },
1657
- {
1658
- "epoch": 0.5067912170505181,
1659
- "grad_norm": 6.561529636383057,
1660
- "learning_rate": 5.132159673685976e-06,
1661
- "loss": 1.614,
1662
- "step": 2360
1663
- },
1664
- {
1665
- "epoch": 0.508938637461749,
1666
- "grad_norm": 17.91114616394043,
1667
- "learning_rate": 5.097385996429992e-06,
1668
- "loss": 2.3969,
1669
- "step": 2370
1670
- },
1671
- {
1672
- "epoch": 0.5110860578729801,
1673
- "grad_norm": 8.510595321655273,
1674
- "learning_rate": 5.062607606291208e-06,
1675
- "loss": 2.0208,
1676
- "step": 2380
1677
- },
1678
- {
1679
- "epoch": 0.5132334782842111,
1680
- "grad_norm": 9.593419075012207,
1681
- "learning_rate": 5.027826186329642e-06,
1682
- "loss": 2.0224,
1683
- "step": 2390
1684
- },
1685
- {
1686
- "epoch": 0.5153808986954421,
1687
- "grad_norm": 9.872200012207031,
1688
- "learning_rate": 4.993043419751933e-06,
1689
- "loss": 1.2689,
1690
- "step": 2400
1691
- },
1692
- {
1693
- "epoch": 0.5175283191066731,
1694
- "grad_norm": 8.374130249023438,
1695
- "learning_rate": 4.958260989829889e-06,
1696
- "loss": 1.2412,
1697
- "step": 2410
1698
- },
1699
- {
1700
- "epoch": 0.5196757395179041,
1701
- "grad_norm": 9.26429271697998,
1702
- "learning_rate": 4.923480579819025e-06,
1703
- "loss": 1.9938,
1704
- "step": 2420
1705
- },
1706
- {
1707
- "epoch": 0.5218231599291351,
1708
- "grad_norm": 9.154571533203125,
1709
- "learning_rate": 4.888703872877108e-06,
1710
- "loss": 1.6627,
1711
- "step": 2430
1712
- },
1713
- {
1714
- "epoch": 0.5239705803403661,
1715
- "grad_norm": 7.144714832305908,
1716
- "learning_rate": 4.853932551982692e-06,
1717
- "loss": 1.675,
1718
- "step": 2440
1719
- },
1720
- {
1721
- "epoch": 0.5261180007515971,
1722
- "grad_norm": 9.676398277282715,
1723
- "learning_rate": 4.8191682998536905e-06,
1724
- "loss": 2.051,
1725
- "step": 2450
1726
- },
1727
- {
1728
- "epoch": 0.5282654211628282,
1729
- "grad_norm": 7.870721817016602,
1730
- "learning_rate": 4.7844127988659204e-06,
1731
- "loss": 1.9328,
1732
- "step": 2460
1733
- },
1734
- {
1735
- "epoch": 0.5304128415740592,
1736
- "grad_norm": 6.848033428192139,
1737
- "learning_rate": 4.749667730971704e-06,
1738
- "loss": 1.7212,
1739
- "step": 2470
1740
- },
1741
- {
1742
- "epoch": 0.5325602619852902,
1743
- "grad_norm": 17.782894134521484,
1744
- "learning_rate": 4.714934777618468e-06,
1745
- "loss": 2.4386,
1746
- "step": 2480
1747
- },
1748
- {
1749
- "epoch": 0.5347076823965212,
1750
- "grad_norm": 17.414901733398438,
1751
- "learning_rate": 4.680215619667364e-06,
1752
- "loss": 2.778,
1753
- "step": 2490
1754
- },
1755
- {
1756
- "epoch": 0.5368551028077522,
1757
- "grad_norm": 9.445879936218262,
1758
- "learning_rate": 4.645511937311934e-06,
1759
- "loss": 2.0429,
1760
- "step": 2500
1761
- },
1762
- {
1763
- "epoch": 0.5390025232189832,
1764
- "grad_norm": 10.506054878234863,
1765
- "learning_rate": 4.610825409996795e-06,
1766
- "loss": 2.454,
1767
- "step": 2510
1768
- },
1769
- {
1770
- "epoch": 0.5411499436302142,
1771
- "grad_norm": 10.331677436828613,
1772
- "learning_rate": 4.576157716336369e-06,
1773
- "loss": 2.0336,
1774
- "step": 2520
1775
- },
1776
- {
1777
- "epoch": 0.5432973640414452,
1778
- "grad_norm": 16.582256317138672,
1779
- "learning_rate": 4.541510534033643e-06,
1780
- "loss": 3.265,
1781
- "step": 2530
1782
- },
1783
- {
1784
- "epoch": 0.5454447844526762,
1785
- "grad_norm": 19.730741500854492,
1786
- "learning_rate": 4.50688553979898e-06,
1787
- "loss": 2.4532,
1788
- "step": 2540
1789
- },
1790
- {
1791
- "epoch": 0.5475922048639073,
1792
- "grad_norm": 19.238706588745117,
1793
- "learning_rate": 4.472284409268976e-06,
1794
- "loss": 2.4381,
1795
- "step": 2550
1796
- },
1797
- {
1798
- "epoch": 0.5497396252751382,
1799
- "grad_norm": 8.645087242126465,
1800
- "learning_rate": 4.437708816925374e-06,
1801
- "loss": 2.0653,
1802
- "step": 2560
1803
- },
1804
- {
1805
- "epoch": 0.5518870456863693,
1806
- "grad_norm": 8.598827362060547,
1807
- "learning_rate": 4.403160436014022e-06,
1808
- "loss": 2.7631,
1809
- "step": 2570
1810
- },
1811
- {
1812
- "epoch": 0.5540344660976002,
1813
- "grad_norm": 17.874841690063477,
1814
- "learning_rate": 4.368640938463909e-06,
1815
- "loss": 3.2379,
1816
- "step": 2580
1817
- },
1818
- {
1819
- "epoch": 0.5561818865088313,
1820
- "grad_norm": 19.613523483276367,
1821
- "learning_rate": 4.334151994806236e-06,
1822
- "loss": 2.3779,
1823
- "step": 2590
1824
- },
1825
- {
1826
- "epoch": 0.5583293069200623,
1827
- "grad_norm": 8.068832397460938,
1828
- "learning_rate": 4.299695274093593e-06,
1829
- "loss": 1.6249,
1830
- "step": 2600
1831
- },
1832
- {
1833
- "epoch": 0.5604767273312933,
1834
- "grad_norm": 9.436213493347168,
1835
- "learning_rate": 4.265272443819175e-06,
1836
- "loss": 2.3983,
1837
- "step": 2610
1838
- },
1839
- {
1840
- "epoch": 0.5626241477425243,
1841
- "grad_norm": 9.81657600402832,
1842
- "learning_rate": 4.23088516983609e-06,
1843
- "loss": 1.9805,
1844
- "step": 2620
1845
- },
1846
- {
1847
- "epoch": 0.5647715681537553,
1848
- "grad_norm": 17.40388298034668,
1849
- "learning_rate": 4.1965351162767344e-06,
1850
- "loss": 2.883,
1851
- "step": 2630
1852
- },
1853
- {
1854
- "epoch": 0.5669189885649863,
1855
- "grad_norm": 8.201501846313477,
1856
- "learning_rate": 4.162223945472271e-06,
1857
- "loss": 2.3845,
1858
- "step": 2640
1859
- },
1860
- {
1861
- "epoch": 0.5690664089762173,
1862
- "grad_norm": 7.743041038513184,
1863
- "learning_rate": 4.1279533178721755e-06,
1864
- "loss": 1.6637,
1865
- "step": 2650
1866
- },
1867
- {
1868
- "epoch": 0.5712138293874484,
1869
- "grad_norm": 9.991127967834473,
1870
- "learning_rate": 4.093724891963882e-06,
1871
- "loss": 1.6559,
1872
- "step": 2660
1873
- },
1874
- {
1875
- "epoch": 0.5733612497986793,
1876
- "grad_norm": 7.750714302062988,
1877
- "learning_rate": 4.059540324192522e-06,
1878
- "loss": 2.4365,
1879
- "step": 2670
1880
- },
1881
- {
1882
- "epoch": 0.5755086702099104,
1883
- "grad_norm": 12.723278999328613,
1884
- "learning_rate": 4.025401268880762e-06,
1885
- "loss": 2.0321,
1886
- "step": 2680
1887
- },
1888
- {
1889
- "epoch": 0.5776560906211413,
1890
- "grad_norm": 9.74173641204834,
1891
- "learning_rate": 3.991309378148746e-06,
1892
- "loss": 2.1192,
1893
- "step": 2690
1894
- },
1895
- {
1896
- "epoch": 0.5798035110323724,
1897
- "grad_norm": 18.213069915771484,
1898
- "learning_rate": 3.957266301834145e-06,
1899
- "loss": 2.3723,
1900
- "step": 2700
1901
- },
1902
- {
1903
- "epoch": 0.5819509314436033,
1904
- "grad_norm": 8.864288330078125,
1905
- "learning_rate": 3.923273687412313e-06,
1906
- "loss": 2.8205,
1907
- "step": 2710
1908
- },
1909
- {
1910
- "epoch": 0.5840983518548344,
1911
- "grad_norm": 8.741086959838867,
1912
- "learning_rate": 3.889333179916552e-06,
1913
- "loss": 2.0988,
1914
- "step": 2720
1915
- },
1916
- {
1917
- "epoch": 0.5862457722660653,
1918
- "grad_norm": 9.561637878417969,
1919
- "learning_rate": 3.855446421858517e-06,
1920
- "loss": 1.9501,
1921
- "step": 2730
1922
- },
1923
- {
1924
- "epoch": 0.5883931926772964,
1925
- "grad_norm": 8.792778015136719,
1926
- "learning_rate": 3.821615053148717e-06,
1927
- "loss": 1.2542,
1928
- "step": 2740
1929
- },
1930
- {
1931
- "epoch": 0.5905406130885275,
1932
- "grad_norm": 8.091140747070312,
1933
- "learning_rate": 3.7878407110171646e-06,
1934
- "loss": 2.0234,
1935
- "step": 2750
1936
- },
1937
- {
1938
- "epoch": 0.5926880334997584,
1939
- "grad_norm": 9.446223258972168,
1940
- "learning_rate": 3.7541250299341243e-06,
1941
- "loss": 2.3535,
1942
- "step": 2760
1943
- },
1944
- {
1945
- "epoch": 0.5948354539109895,
1946
- "grad_norm": 7.457852363586426,
1947
- "learning_rate": 3.7204696415310377e-06,
1948
- "loss": 2.379,
1949
- "step": 2770
1950
- },
1951
- {
1952
- "epoch": 0.5969828743222204,
1953
- "grad_norm": 10.4889497756958,
1954
- "learning_rate": 3.6868761745215474e-06,
1955
- "loss": 2.455,
1956
- "step": 2780
1957
- },
1958
- {
1959
- "epoch": 0.5991302947334515,
1960
- "grad_norm": 7.388014316558838,
1961
- "learning_rate": 3.653346254622683e-06,
1962
- "loss": 1.6059,
1963
- "step": 2790
1964
- },
1965
- {
1966
- "epoch": 0.6012777151446824,
1967
- "grad_norm": 8.156157493591309,
1968
- "learning_rate": 3.6198815044761847e-06,
1969
- "loss": 2.8084,
1970
- "step": 2800
1971
- },
1972
- {
1973
- "epoch": 0.6034251355559135,
1974
- "grad_norm": 10.154397010803223,
1975
- "learning_rate": 3.586483543569977e-06,
1976
- "loss": 2.0311,
1977
- "step": 2810
1978
- },
1979
- {
1980
- "epoch": 0.6055725559671444,
1981
- "grad_norm": 9.225127220153809,
1982
- "learning_rate": 3.5531539881597967e-06,
1983
- "loss": 2.4153,
1984
- "step": 2820
1985
- },
1986
- {
1987
- "epoch": 0.6077199763783755,
1988
- "grad_norm": 17.420804977416992,
1989
- "learning_rate": 3.519894451190976e-06,
1990
- "loss": 2.7263,
1991
- "step": 2830
1992
- },
1993
- {
1994
- "epoch": 0.6098673967896064,
1995
- "grad_norm": 7.618463039398193,
1996
- "learning_rate": 3.4867065422203885e-06,
1997
- "loss": 1.9863,
1998
- "step": 2840
1999
- },
2000
- {
2001
- "epoch": 0.6120148172008375,
2002
- "grad_norm": 7.851476192474365,
2003
- "learning_rate": 3.4535918673385456e-06,
2004
- "loss": 1.6099,
2005
- "step": 2850
2006
- },
2007
- {
2008
- "epoch": 0.6141622376120685,
2009
- "grad_norm": 17.60761260986328,
2010
- "learning_rate": 3.420552029091886e-06,
2011
- "loss": 2.0744,
2012
- "step": 2860
2013
- },
2014
- {
2015
- "epoch": 0.6163096580232995,
2016
- "grad_norm": 7.814233779907227,
2017
- "learning_rate": 3.3875886264052155e-06,
2018
- "loss": 1.9523,
2019
- "step": 2870
2020
- },
2021
- {
2022
- "epoch": 0.6184570784345306,
2023
- "grad_norm": 8.906336784362793,
2024
- "learning_rate": 3.354703254504328e-06,
2025
- "loss": 2.4133,
2026
- "step": 2880
2027
- },
2028
- {
2029
- "epoch": 0.6206044988457615,
2030
- "grad_norm": 10.231346130371094,
2031
- "learning_rate": 3.3218975048388067e-06,
2032
- "loss": 1.644,
2033
- "step": 2890
2034
- },
2035
- {
2036
- "epoch": 0.6227519192569926,
2037
- "grad_norm": 8.996302604675293,
2038
- "learning_rate": 3.2891729650050096e-06,
2039
- "loss": 2.3329,
2040
- "step": 2900
2041
- },
2042
- {
2043
- "epoch": 0.6248993396682235,
2044
- "grad_norm": 8.609686851501465,
2045
- "learning_rate": 3.2565312186692403e-06,
2046
- "loss": 1.6096,
2047
- "step": 2910
2048
- },
2049
- {
2050
- "epoch": 0.6270467600794546,
2051
- "grad_norm": 7.7054338455200195,
2052
- "learning_rate": 3.2239738454911057e-06,
2053
- "loss": 2.4928,
2054
- "step": 2920
2055
- },
2056
- {
2057
- "epoch": 0.6291941804906855,
2058
- "grad_norm": 7.8000383377075195,
2059
- "learning_rate": 3.191502421047068e-06,
2060
- "loss": 2.0236,
2061
- "step": 2930
2062
- },
2063
- {
2064
- "epoch": 0.6313416009019166,
2065
- "grad_norm": 17.71788215637207,
2066
- "learning_rate": 3.1591185167542047e-06,
2067
- "loss": 2.0003,
2068
- "step": 2940
2069
- },
2070
- {
2071
- "epoch": 0.6334890213131475,
2072
- "grad_norm": 10.549698829650879,
2073
- "learning_rate": 3.1268236997941535e-06,
2074
- "loss": 1.2635,
2075
- "step": 2950
2076
- },
2077
- {
2078
- "epoch": 0.6356364417243786,
2079
- "grad_norm": 18.817182540893555,
2080
- "learning_rate": 3.0946195330372754e-06,
2081
- "loss": 3.523,
2082
- "step": 2960
2083
- },
2084
- {
2085
- "epoch": 0.6377838621356096,
2086
- "grad_norm": 7.883931636810303,
2087
- "learning_rate": 3.0625075749670195e-06,
2088
- "loss": 2.2924,
2089
- "step": 2970
2090
- },
2091
- {
2092
- "epoch": 0.6399312825468406,
2093
- "grad_norm": 8.429082870483398,
2094
- "learning_rate": 3.0304893796044988e-06,
2095
- "loss": 1.6421,
2096
- "step": 2980
2097
- },
2098
- {
2099
- "epoch": 0.6420787029580716,
2100
- "grad_norm": 8.13762092590332,
2101
- "learning_rate": 2.9985664964332904e-06,
2102
- "loss": 1.9695,
2103
- "step": 2990
2104
- },
2105
- {
2106
- "epoch": 0.6442261233693026,
2107
- "grad_norm": 16.98133659362793,
2108
- "learning_rate": 2.966740470324451e-06,
2109
- "loss": 3.2073,
2110
- "step": 3000
2111
- },
2112
- {
2113
- "epoch": 0.6463735437805337,
2114
- "grad_norm": 8.005885124206543,
2115
- "learning_rate": 2.9350128414617483e-06,
2116
- "loss": 2.0849,
2117
- "step": 3010
2118
- },
2119
- {
2120
- "epoch": 0.6485209641917646,
2121
- "grad_norm": 18.396974563598633,
2122
- "learning_rate": 2.903385145267129e-06,
2123
- "loss": 2.4081,
2124
- "step": 3020
2125
- },
2126
- {
2127
- "epoch": 0.6506683846029957,
2128
- "grad_norm": 11.773588180541992,
2129
- "learning_rate": 2.871858912326414e-06,
2130
- "loss": 1.6188,
2131
- "step": 3030
2132
- },
2133
- {
2134
- "epoch": 0.6528158050142266,
2135
- "grad_norm": 12.255626678466797,
2136
- "learning_rate": 2.8404356683152256e-06,
2137
- "loss": 2.4209,
2138
- "step": 3040
2139
- },
2140
- {
2141
- "epoch": 0.6549632254254577,
2142
- "grad_norm": 7.566231727600098,
2143
- "learning_rate": 2.8091169339251644e-06,
2144
- "loss": 1.6243,
2145
- "step": 3050
2146
- },
2147
- {
2148
- "epoch": 0.6571106458366887,
2149
- "grad_norm": 8.969308853149414,
2150
- "learning_rate": 2.777904224790197e-06,
2151
- "loss": 1.671,
2152
- "step": 3060
2153
- },
2154
- {
2155
- "epoch": 0.6592580662479197,
2156
- "grad_norm": 7.9166364669799805,
2157
- "learning_rate": 2.746799051413325e-06,
2158
- "loss": 2.461,
2159
- "step": 3070
2160
- },
2161
- {
2162
- "epoch": 0.6614054866591507,
2163
- "grad_norm": 16.861021041870117,
2164
- "learning_rate": 2.715802919093484e-06,
2165
- "loss": 1.9931,
2166
- "step": 3080
2167
- },
2168
- {
2169
- "epoch": 0.6635529070703817,
2170
- "grad_norm": 18.910966873168945,
2171
- "learning_rate": 2.684917327852691e-06,
2172
- "loss": 2.483,
2173
- "step": 3090
2174
- },
2175
- {
2176
- "epoch": 0.6657003274816127,
2177
- "grad_norm": 7.5983686447143555,
2178
- "learning_rate": 2.654143772363455e-06,
2179
- "loss": 1.9728,
2180
- "step": 3100
2181
- },
2182
- {
2183
- "epoch": 0.6678477478928437,
2184
- "grad_norm": 8.704245567321777,
2185
- "learning_rate": 2.623483741876443e-06,
2186
- "loss": 1.5872,
2187
- "step": 3110
2188
- },
2189
- {
2190
- "epoch": 0.6699951683040747,
2191
- "grad_norm": 10.111562728881836,
2192
- "learning_rate": 2.5929387201484133e-06,
2193
- "loss": 1.5902,
2194
- "step": 3120
2195
- },
2196
- {
2197
- "epoch": 0.6721425887153057,
2198
- "grad_norm": 18.483325958251953,
2199
- "learning_rate": 2.562510185370407e-06,
2200
- "loss": 2.751,
2201
- "step": 3130
2202
- },
2203
- {
2204
- "epoch": 0.6742900091265367,
2205
- "grad_norm": 7.957283020019531,
2206
- "learning_rate": 2.5321996100962163e-06,
2207
- "loss": 2.3621,
2208
- "step": 3140
2209
- },
2210
- {
2211
- "epoch": 0.6764374295377678,
2212
- "grad_norm": 7.387941837310791,
2213
- "learning_rate": 2.502008461171114e-06,
2214
- "loss": 1.5594,
2215
- "step": 3150
2216
- },
2217
- {
2218
- "epoch": 0.6785848499489988,
2219
- "grad_norm": 8.454072952270508,
2220
- "learning_rate": 2.4719381996608748e-06,
2221
- "loss": 3.2272,
2222
- "step": 3160
2223
- },
2224
- {
2225
- "epoch": 0.6807322703602298,
2226
- "grad_norm": 8.459441184997559,
2227
- "learning_rate": 2.4419902807810707e-06,
2228
- "loss": 1.5914,
2229
- "step": 3170
2230
- },
2231
- {
2232
- "epoch": 0.6828796907714608,
2233
- "grad_norm": 8.403077125549316,
2234
- "learning_rate": 2.412166153826639e-06,
2235
- "loss": 1.6618,
2236
- "step": 3180
2237
- },
2238
- {
2239
- "epoch": 0.6850271111826918,
2240
- "grad_norm": 8.397653579711914,
2241
- "learning_rate": 2.382467262101751e-06,
2242
- "loss": 1.6576,
2243
- "step": 3190
2244
- },
2245
- {
2246
- "epoch": 0.6871745315939228,
2247
- "grad_norm": 8.854833602905273,
2248
- "learning_rate": 2.352895042849965e-06,
2249
- "loss": 2.0303,
2250
- "step": 3200
2251
- },
2252
- {
2253
- "epoch": 0.6893219520051538,
2254
- "grad_norm": 17.544239044189453,
2255
- "learning_rate": 2.3234509271846683e-06,
2256
- "loss": 3.0549,
2257
- "step": 3210
2258
- },
2259
- {
2260
- "epoch": 0.6914693724163848,
2261
- "grad_norm": 20.67961883544922,
2262
- "learning_rate": 2.294136340019826e-06,
2263
- "loss": 2.3566,
2264
- "step": 3220
2265
- },
2266
- {
2267
- "epoch": 0.6936167928276158,
2268
- "grad_norm": 18.9661922454834,
2269
- "learning_rate": 2.264952700001022e-06,
2270
- "loss": 2.746,
2271
- "step": 3230
2272
- },
2273
- {
2274
- "epoch": 0.6957642132388469,
2275
- "grad_norm": 7.390661716461182,
2276
- "learning_rate": 2.2359014194367986e-06,
2277
- "loss": 1.971,
2278
- "step": 3240
2279
- },
2280
- {
2281
- "epoch": 0.6979116336500778,
2282
- "grad_norm": 6.957170486450195,
2283
- "learning_rate": 2.20698390423032e-06,
2284
- "loss": 1.2324,
2285
- "step": 3250
2286
- },
2287
- {
2288
- "epoch": 0.7000590540613089,
2289
- "grad_norm": 17.50217628479004,
2290
- "learning_rate": 2.17820155381133e-06,
2291
- "loss": 2.3776,
2292
- "step": 3260
2293
- },
2294
- {
2295
- "epoch": 0.7022064744725398,
2296
- "grad_norm": 8.914453506469727,
2297
- "learning_rate": 2.14955576106843e-06,
2298
- "loss": 2.3974,
2299
- "step": 3270
2300
- },
2301
- {
2302
- "epoch": 0.7043538948837709,
2303
- "grad_norm": 18.855152130126953,
2304
- "learning_rate": 2.1210479122816646e-06,
2305
- "loss": 1.6601,
2306
- "step": 3280
2307
- },
2308
- {
2309
- "epoch": 0.7065013152950019,
2310
- "grad_norm": 18.482494354248047,
2311
- "learning_rate": 2.0926793870554457e-06,
2312
- "loss": 1.984,
2313
- "step": 3290
2314
- },
2315
- {
2316
- "epoch": 0.7086487357062329,
2317
- "grad_norm": 8.10519027709961,
2318
- "learning_rate": 2.0644515582517803e-06,
2319
- "loss": 1.9964,
2320
- "step": 3300
2321
- },
2322
- {
2323
- "epoch": 0.7107961561174639,
2324
- "grad_norm": 19.076732635498047,
2325
- "learning_rate": 2.0363657919238357e-06,
2326
- "loss": 2.7817,
2327
- "step": 3310
2328
- },
2329
- {
2330
- "epoch": 0.7129435765286949,
2331
- "grad_norm": 18.62787628173828,
2332
- "learning_rate": 2.0084234472498274e-06,
2333
- "loss": 2.3052,
2334
- "step": 3320
2335
- },
2336
- {
2337
- "epoch": 0.715090996939926,
2338
- "grad_norm": 8.185526847839355,
2339
- "learning_rate": 1.9806258764672488e-06,
2340
- "loss": 2.2661,
2341
- "step": 3330
2342
- },
2343
- {
2344
- "epoch": 0.7172384173511569,
2345
- "grad_norm": 8.022353172302246,
2346
- "learning_rate": 1.952974424807425e-06,
2347
- "loss": 2.0014,
2348
- "step": 3340
2349
- },
2350
- {
2351
- "epoch": 0.719385837762388,
2352
- "grad_norm": 10.178295135498047,
2353
- "learning_rate": 1.9254704304304174e-06,
2354
- "loss": 1.5894,
2355
- "step": 3350
2356
- },
2357
- {
2358
- "epoch": 0.7215332581736189,
2359
- "grad_norm": 8.717124938964844,
2360
- "learning_rate": 1.898115224360263e-06,
2361
- "loss": 1.1596,
2362
- "step": 3360
2363
- },
2364
- {
2365
- "epoch": 0.72368067858485,
2366
- "grad_norm": 8.687152862548828,
2367
- "learning_rate": 1.870910130420555e-06,
2368
- "loss": 2.0217,
2369
- "step": 3370
2370
- },
2371
- {
2372
- "epoch": 0.7258280989960809,
2373
- "grad_norm": 8.655681610107422,
2374
- "learning_rate": 1.84385646517039e-06,
2375
- "loss": 1.557,
2376
- "step": 3380
2377
- },
2378
- {
2379
- "epoch": 0.727975519407312,
2380
- "grad_norm": 18.08987808227539,
2381
- "learning_rate": 1.8169555378406456e-06,
2382
- "loss": 2.3207,
2383
- "step": 3390
2384
- },
2385
- {
2386
- "epoch": 0.7301229398185429,
2387
- "grad_norm": 8.34890079498291,
2388
- "learning_rate": 1.7902086502706256e-06,
2389
- "loss": 1.9965,
2390
- "step": 3400
2391
- },
2392
- {
2393
- "epoch": 0.732270360229774,
2394
- "grad_norm": 7.2736921310424805,
2395
- "learning_rate": 1.7636170968450533e-06,
2396
- "loss": 1.5618,
2397
- "step": 3410
2398
- },
2399
- {
2400
- "epoch": 0.734417780641005,
2401
- "grad_norm": 7.780071258544922,
2402
- "learning_rate": 1.7371821644314392e-06,
2403
- "loss": 2.3343,
2404
- "step": 3420
2405
- },
2406
- {
2407
- "epoch": 0.736565201052236,
2408
- "grad_norm": 8.583059310913086,
2409
- "learning_rate": 1.710905132317801e-06,
2410
- "loss": 2.7489,
2411
- "step": 3430
2412
- },
2413
- {
2414
- "epoch": 0.7387126214634671,
2415
- "grad_norm": 8.453703880310059,
2416
- "learning_rate": 1.6847872721507525e-06,
2417
- "loss": 2.0472,
2418
- "step": 3440
2419
- },
2420
- {
2421
- "epoch": 0.740860041874698,
2422
- "grad_norm": 8.093005180358887,
2423
- "learning_rate": 1.658829847873965e-06,
2424
- "loss": 1.6072,
2425
- "step": 3450
2426
- },
2427
- {
2428
- "epoch": 0.7430074622859291,
2429
- "grad_norm": 17.80980110168457,
2430
- "learning_rate": 1.633034115667001e-06,
2431
- "loss": 1.9875,
2432
- "step": 3460
2433
- },
2434
- {
2435
- "epoch": 0.74515488269716,
2436
- "grad_norm": 19.054296493530273,
2437
- "learning_rate": 1.6074013238845214e-06,
2438
- "loss": 2.7245,
2439
- "step": 3470
2440
- },
2441
- {
2442
- "epoch": 0.7473023031083911,
2443
- "grad_norm": 7.945253372192383,
2444
- "learning_rate": 1.5819327129958762e-06,
2445
- "loss": 2.3126,
2446
- "step": 3480
2447
- },
2448
- {
2449
- "epoch": 0.749449723519622,
2450
- "grad_norm": 8.40179443359375,
2451
- "learning_rate": 1.5566295155250644e-06,
2452
- "loss": 2.3687,
2453
- "step": 3490
2454
- },
2455
- {
2456
- "epoch": 0.7515971439308531,
2457
- "grad_norm": 8.832183837890625,
2458
- "learning_rate": 1.5314929559910985e-06,
2459
- "loss": 1.6223,
2460
- "step": 3500
2461
- },
2462
- {
2463
- "epoch": 0.753744564342084,
2464
- "grad_norm": 9.67546558380127,
2465
- "learning_rate": 1.506524250848741e-06,
2466
- "loss": 2.3248,
2467
- "step": 3510
2468
- },
2469
- {
2470
- "epoch": 0.7558919847533151,
2471
- "grad_norm": 7.6303558349609375,
2472
- "learning_rate": 1.4817246084296327e-06,
2473
- "loss": 1.2378,
2474
- "step": 3520
2475
- },
2476
- {
2477
- "epoch": 0.758039405164546,
2478
- "grad_norm": 10.63131046295166,
2479
- "learning_rate": 1.457095228883822e-06,
2480
- "loss": 1.1994,
2481
- "step": 3530
2482
- },
2483
- {
2484
- "epoch": 0.7601868255757771,
2485
- "grad_norm": 11.350310325622559,
2486
- "learning_rate": 1.4326373041216774e-06,
2487
- "loss": 1.5317,
2488
- "step": 3540
2489
- },
2490
- {
2491
- "epoch": 0.762334245987008,
2492
- "grad_norm": 9.425748825073242,
2493
- "learning_rate": 1.4083520177562154e-06,
2494
- "loss": 1.5042,
2495
- "step": 3550
2496
- },
2497
- {
2498
- "epoch": 0.7644816663982391,
2499
- "grad_norm": 18.74260902404785,
2500
- "learning_rate": 1.3842405450458158e-06,
2501
- "loss": 2.3408,
2502
- "step": 3560
2503
- },
2504
- {
2505
- "epoch": 0.7666290868094702,
2506
- "grad_norm": 8.119706153869629,
2507
- "learning_rate": 1.3603040528373467e-06,
2508
- "loss": 2.3104,
2509
- "step": 3570
2510
- },
2511
- {
2512
- "epoch": 0.7687765072207011,
2513
- "grad_norm": 10.3068208694458,
2514
- "learning_rate": 1.336543699509698e-06,
2515
- "loss": 2.5002,
2516
- "step": 3580
2517
- },
2518
- {
2519
- "epoch": 0.7709239276319322,
2520
- "grad_norm": 17.977724075317383,
2521
- "learning_rate": 1.312960634917721e-06,
2522
- "loss": 2.6457,
2523
- "step": 3590
2524
- },
2525
- {
2526
- "epoch": 0.7730713480431631,
2527
- "grad_norm": 9.48388957977295,
2528
- "learning_rate": 1.2895560003365837e-06,
2529
- "loss": 1.9769,
2530
- "step": 3600
2531
- },
2532
- {
2533
- "epoch": 0.7752187684543942,
2534
- "grad_norm": 8.272026062011719,
2535
- "learning_rate": 1.2663309284065407e-06,
2536
- "loss": 2.9297,
2537
- "step": 3610
2538
- },
2539
- {
2540
- "epoch": 0.7773661888656251,
2541
- "grad_norm": 8.88364315032959,
2542
- "learning_rate": 1.2432865430781166e-06,
2543
- "loss": 1.6369,
2544
- "step": 3620
2545
- },
2546
- {
2547
- "epoch": 0.7795136092768562,
2548
- "grad_norm": 7.110147476196289,
2549
- "learning_rate": 1.2204239595577195e-06,
2550
- "loss": 1.5633,
2551
- "step": 3630
2552
- },
2553
- {
2554
- "epoch": 0.7816610296880872,
2555
- "grad_norm": 8.931872367858887,
2556
- "learning_rate": 1.1977442842536685e-06,
2557
- "loss": 1.5733,
2558
- "step": 3640
2559
- },
2560
- {
2561
- "epoch": 0.7838084500993182,
2562
- "grad_norm": 8.303011894226074,
2563
- "learning_rate": 1.1752486147226505e-06,
2564
- "loss": 1.9338,
2565
- "step": 3650
2566
- },
2567
- {
2568
- "epoch": 0.7859558705105492,
2569
- "grad_norm": 7.458680629730225,
2570
- "learning_rate": 1.1529380396166074e-06,
2571
- "loss": 3.1121,
2572
- "step": 3660
2573
- },
2574
- {
2575
- "epoch": 0.7881032909217802,
2576
- "grad_norm": 7.7846598625183105,
2577
- "learning_rate": 1.1308136386300455e-06,
2578
- "loss": 1.5423,
2579
- "step": 3670
2580
- },
2581
- {
2582
- "epoch": 0.7902507113330112,
2583
- "grad_norm": 10.858473777770996,
2584
- "learning_rate": 1.1088764824477938e-06,
2585
- "loss": 1.2332,
2586
- "step": 3680
2587
- },
2588
- {
2589
- "epoch": 0.7923981317442422,
2590
- "grad_norm": 10.06197738647461,
2591
- "learning_rate": 1.0871276326931845e-06,
2592
- "loss": 1.9164,
2593
- "step": 3690
2594
- },
2595
- {
2596
- "epoch": 0.7945455521554733,
2597
- "grad_norm": 7.83908224105835,
2598
- "learning_rate": 1.0655681418766772e-06,
2599
- "loss": 1.9849,
2600
- "step": 3700
2601
- },
2602
- {
2603
- "epoch": 0.7966929725667042,
2604
- "grad_norm": 8.16726303100586,
2605
- "learning_rate": 1.0441990533449247e-06,
2606
- "loss": 1.6056,
2607
- "step": 3710
2608
- },
2609
- {
2610
- "epoch": 0.7988403929779353,
2611
- "grad_norm": 17.369827270507812,
2612
- "learning_rate": 1.0230214012302807e-06,
2613
- "loss": 2.7502,
2614
- "step": 3720
2615
- },
2616
- {
2617
- "epoch": 0.8009878133891662,
2618
- "grad_norm": 9.131800651550293,
2619
- "learning_rate": 1.0020362104007558e-06,
2620
- "loss": 1.246,
2621
- "step": 3730
2622
- },
2623
- {
2624
- "epoch": 0.8031352338003973,
2625
- "grad_norm": 17.537370681762695,
2626
- "learning_rate": 9.812444964104195e-07,
2627
- "loss": 1.5823,
2628
- "step": 3740
2629
- },
2630
- {
2631
- "epoch": 0.8052826542116283,
2632
- "grad_norm": 8.849536895751953,
2633
- "learning_rate": 9.60647265450249e-07,
2634
- "loss": 1.6186,
2635
- "step": 3750
2636
- },
2637
- {
2638
- "epoch": 0.8074300746228593,
2639
- "grad_norm": 8.354120254516602,
2640
- "learning_rate": 9.402455142994443e-07,
2641
- "loss": 2.3345,
2642
- "step": 3760
2643
- },
2644
- {
2645
- "epoch": 0.8095774950340903,
2646
- "grad_norm": 8.479084968566895,
2647
- "learning_rate": 9.200402302771843e-07,
2648
- "loss": 2.3736,
2649
- "step": 3770
2650
- },
2651
- {
2652
- "epoch": 0.8117249154453213,
2653
- "grad_norm": 8.048047065734863,
2654
- "learning_rate": 9.000323911948483e-07,
2655
- "loss": 1.6527,
2656
- "step": 3780
2657
- },
2658
- {
2659
- "epoch": 0.8138723358565523,
2660
- "grad_norm": 9.944975852966309,
2661
- "learning_rate": 8.802229653086975e-07,
2662
- "loss": 2.3705,
2663
- "step": 3790
2664
- },
2665
- {
2666
- "epoch": 0.8160197562677833,
2667
- "grad_norm": 19.067068099975586,
2668
- "learning_rate": 8.60612911273011e-07,
2669
- "loss": 2.3448,
2670
- "step": 3800
2671
- },
2672
- {
2673
- "epoch": 0.8181671766790143,
2674
- "grad_norm": 11.265616416931152,
2675
- "learning_rate": 8.412031780937025e-07,
2676
- "loss": 1.5598,
2677
- "step": 3810
2678
- },
2679
- {
2680
- "epoch": 0.8203145970902453,
2681
- "grad_norm": 7.994141101837158,
2682
- "learning_rate": 8.219947050823862e-07,
2683
- "loss": 1.5608,
2684
- "step": 3820
2685
- },
2686
- {
2687
- "epoch": 0.8224620175014764,
2688
- "grad_norm": 10.012679100036621,
2689
- "learning_rate": 8.029884218109246e-07,
2690
- "loss": 1.5939,
2691
- "step": 3830
2692
- },
2693
- {
2694
- "epoch": 0.8246094379127074,
2695
- "grad_norm": 7.568084716796875,
2696
- "learning_rate": 7.841852480664414e-07,
2697
- "loss": 1.5617,
2698
- "step": 3840
2699
- },
2700
- {
2701
- "epoch": 0.8267568583239384,
2702
- "grad_norm": 8.215276718139648,
2703
- "learning_rate": 7.655860938068071e-07,
2704
- "loss": 1.5861,
2705
- "step": 3850
2706
- },
2707
- {
2708
- "epoch": 0.8289042787351694,
2709
- "grad_norm": 9.24499797821045,
2710
- "learning_rate": 7.471918591166078e-07,
2711
- "loss": 2.6894,
2712
- "step": 3860
2713
- },
2714
- {
2715
- "epoch": 0.8310516991464004,
2716
- "grad_norm": 17.422555923461914,
2717
- "learning_rate": 7.290034341635838e-07,
2718
- "loss": 1.9686,
2719
- "step": 3870
2720
- },
2721
- {
2722
- "epoch": 0.8331991195576314,
2723
- "grad_norm": 17.70247459411621,
2724
- "learning_rate": 7.110216991555457e-07,
2725
- "loss": 2.3266,
2726
- "step": 3880
2727
- },
2728
- {
2729
- "epoch": 0.8353465399688624,
2730
- "grad_norm": 8.571028709411621,
2731
- "learning_rate": 6.932475242977899e-07,
2732
- "loss": 1.5555,
2733
- "step": 3890
2734
- },
2735
- {
2736
- "epoch": 0.8374939603800934,
2737
- "grad_norm": 8.366564750671387,
2738
- "learning_rate": 6.756817697509755e-07,
2739
- "loss": 1.5639,
2740
- "step": 3900
2741
- },
2742
- {
2743
- "epoch": 0.8396413807913244,
2744
- "grad_norm": 7.642975807189941,
2745
- "learning_rate": 6.583252855895012e-07,
2746
- "loss": 1.1573,
2747
- "step": 3910
2748
- },
2749
- {
2750
- "epoch": 0.8417888012025554,
2751
- "grad_norm": 9.669418334960938,
2752
- "learning_rate": 6.411789117603701e-07,
2753
- "loss": 2.3294,
2754
- "step": 3920
2755
- },
2756
- {
2757
- "epoch": 0.8439362216137865,
2758
- "grad_norm": 8.85915470123291,
2759
- "learning_rate": 6.242434780425333e-07,
2760
- "loss": 1.5266,
2761
- "step": 3930
2762
- },
2763
- {
2764
- "epoch": 0.8460836420250174,
2765
- "grad_norm": 8.256942749023438,
2766
- "learning_rate": 6.075198040067432e-07,
2767
- "loss": 1.5556,
2768
- "step": 3940
2769
- },
2770
- {
2771
- "epoch": 0.8482310624362485,
2772
- "grad_norm": 8.303900718688965,
2773
- "learning_rate": 5.910086989758862e-07,
2774
- "loss": 1.9915,
2775
- "step": 3950
2776
- },
2777
- {
2778
- "epoch": 0.8503784828474794,
2779
- "grad_norm": 8.628076553344727,
2780
- "learning_rate": 5.747109619858176e-07,
2781
- "loss": 1.2209,
2782
- "step": 3960
2783
- },
2784
- {
2785
- "epoch": 0.8525259032587105,
2786
- "grad_norm": 8.535489082336426,
2787
- "learning_rate": 5.586273817466891e-07,
2788
- "loss": 2.7191,
2789
- "step": 3970
2790
- },
2791
- {
2792
- "epoch": 0.8546733236699415,
2793
- "grad_norm": 10.173384666442871,
2794
- "learning_rate": 5.427587366047893e-07,
2795
- "loss": 2.3468,
2796
- "step": 3980
2797
- },
2798
- {
2799
- "epoch": 0.8568207440811725,
2800
- "grad_norm": 8.549752235412598,
2801
- "learning_rate": 5.27105794504868e-07,
2802
- "loss": 2.3646,
2803
- "step": 3990
2804
- },
2805
- {
2806
- "epoch": 0.8589681644924035,
2807
- "grad_norm": 9.579891204833984,
2808
- "learning_rate": 5.11669312952977e-07,
2809
- "loss": 1.54,
2810
- "step": 4000
2811
- },
2812
- {
2813
- "epoch": 0.8611155849036345,
2814
- "grad_norm": 8.296952247619629,
2815
- "learning_rate": 4.964500389798066e-07,
2816
- "loss": 1.9316,
2817
- "step": 4010
2818
- },
2819
- {
2820
- "epoch": 0.8632630053148656,
2821
- "grad_norm": 16.096107482910156,
2822
- "learning_rate": 4.814487091045405e-07,
2823
- "loss": 2.6136,
2824
- "step": 4020
2825
- },
2826
- {
2827
- "epoch": 0.8654104257260965,
2828
- "grad_norm": 8.543082237243652,
2829
- "learning_rate": 4.666660492992092e-07,
2830
- "loss": 1.5991,
2831
- "step": 4030
2832
- },
2833
- {
2834
- "epoch": 0.8675578461373276,
2835
- "grad_norm": 7.935061931610107,
2836
- "learning_rate": 4.521027749535578e-07,
2837
- "loss": 1.921,
2838
- "step": 4040
2839
- },
2840
- {
2841
- "epoch": 0.8697052665485585,
2842
- "grad_norm": 9.394523620605469,
2843
- "learning_rate": 4.377595908404225e-07,
2844
- "loss": 1.5151,
2845
- "step": 4050
2846
- },
2847
- {
2848
- "epoch": 0.8718526869597896,
2849
- "grad_norm": 8.564217567443848,
2850
- "learning_rate": 4.2363719108163113e-07,
2851
- "loss": 1.1554,
2852
- "step": 4060
2853
- },
2854
- {
2855
- "epoch": 0.8740001073710205,
2856
- "grad_norm": 10.94825267791748,
2857
- "learning_rate": 4.097362591144055e-07,
2858
- "loss": 1.1995,
2859
- "step": 4070
2860
- },
2861
- {
2862
- "epoch": 0.8761475277822516,
2863
- "grad_norm": 17.23007583618164,
2864
- "learning_rate": 3.960574676582901e-07,
2865
- "loss": 2.391,
2866
- "step": 4080
2867
- },
2868
- {
2869
- "epoch": 0.8782949481934825,
2870
- "grad_norm": 9.983611106872559,
2871
- "learning_rate": 3.8260147868259713e-07,
2872
- "loss": 3.2049,
2873
- "step": 4090
2874
- },
2875
- {
2876
- "epoch": 0.8804423686047136,
2877
- "grad_norm": 7.7599921226501465,
2878
- "learning_rate": 3.693689433743658e-07,
2879
- "loss": 2.2062,
2880
- "step": 4100
2881
- },
2882
- {
2883
- "epoch": 0.8825897890159446,
2884
- "grad_norm": 8.449335098266602,
2885
- "learning_rate": 3.563605021068578e-07,
2886
- "loss": 1.9016,
2887
- "step": 4110
2888
- },
2889
- {
2890
- "epoch": 0.8847372094271756,
2891
- "grad_norm": 18.595163345336914,
2892
- "learning_rate": 3.4357678440856136e-07,
2893
- "loss": 2.035,
2894
- "step": 4120
2895
- },
2896
- {
2897
- "epoch": 0.8868846298384067,
2898
- "grad_norm": 8.655158996582031,
2899
- "learning_rate": 3.3101840893272786e-07,
2900
- "loss": 1.6087,
2901
- "step": 4130
2902
- },
2903
- {
2904
- "epoch": 0.8890320502496376,
2905
- "grad_norm": 8.49612045288086,
2906
- "learning_rate": 3.186859834274292e-07,
2907
- "loss": 1.5791,
2908
- "step": 4140
2909
- },
2910
- {
2911
- "epoch": 0.8911794706608687,
2912
- "grad_norm": 8.00724983215332,
2913
- "learning_rate": 3.065801047061517e-07,
2914
- "loss": 1.2228,
2915
- "step": 4150
2916
- },
2917
- {
2918
- "epoch": 0.8933268910720996,
2919
- "grad_norm": 18.299640655517578,
2920
- "learning_rate": 2.947013586189124e-07,
2921
- "loss": 2.3075,
2922
- "step": 4160
2923
- },
2924
- {
2925
- "epoch": 0.8954743114833307,
2926
- "grad_norm": 9.55125904083252,
2927
- "learning_rate": 2.83050320023906e-07,
2928
- "loss": 2.684,
2929
- "step": 4170
2930
- },
2931
- {
2932
- "epoch": 0.8976217318945616,
2933
- "grad_norm": 9.116815567016602,
2934
- "learning_rate": 2.7162755275968513e-07,
2935
- "loss": 1.9426,
2936
- "step": 4180
2937
- },
2938
- {
2939
- "epoch": 0.8997691523057927,
2940
- "grad_norm": 12.08741283416748,
2941
- "learning_rate": 2.604336096178767e-07,
2942
- "loss": 1.9566,
2943
- "step": 4190
2944
- },
2945
- {
2946
- "epoch": 0.9019165727170236,
2947
- "grad_norm": 17.83538818359375,
2948
- "learning_rate": 2.4946903231642727e-07,
2949
- "loss": 2.3625,
2950
- "step": 4200
2951
- },
2952
- {
2953
- "epoch": 0.9040639931282547,
2954
- "grad_norm": 8.699211120605469,
2955
- "learning_rate": 2.3873435147338975e-07,
2956
- "loss": 1.6144,
2957
- "step": 4210
2958
- },
2959
- {
2960
- "epoch": 0.9062114135394856,
2961
- "grad_norm": 11.926467895507812,
2962
- "learning_rate": 2.2823008658124425e-07,
2963
- "loss": 1.6301,
2964
- "step": 4220
2965
- },
2966
- {
2967
- "epoch": 0.9083588339507167,
2968
- "grad_norm": 18.642906188964844,
2969
- "learning_rate": 2.1795674598175277e-07,
2970
- "loss": 2.007,
2971
- "step": 4230
2972
- },
2973
- {
2974
- "epoch": 0.9105062543619478,
2975
- "grad_norm": 8.510333061218262,
2976
- "learning_rate": 2.0791482684136833e-07,
2977
- "loss": 2.3849,
2978
- "step": 4240
2979
- },
2980
- {
2981
- "epoch": 0.9126536747731787,
2982
- "grad_norm": 8.797530174255371,
2983
- "learning_rate": 1.9810481512716638e-07,
2984
- "loss": 2.2797,
2985
- "step": 4250
2986
- },
2987
- {
2988
- "epoch": 0.9148010951844098,
2989
- "grad_norm": 9.010210990905762,
2990
- "learning_rate": 1.885271855833315e-07,
2991
- "loss": 1.9205,
2992
- "step": 4260
2993
- },
2994
- {
2995
- "epoch": 0.9169485155956407,
2996
- "grad_norm": 6.828660488128662,
2997
- "learning_rate": 1.7918240170818225e-07,
2998
- "loss": 1.9204,
2999
- "step": 4270
3000
- },
3001
- {
3002
- "epoch": 0.9190959360068718,
3003
- "grad_norm": 17.962793350219727,
3004
- "learning_rate": 1.7007091573173818e-07,
3005
- "loss": 2.3651,
3006
- "step": 4280
3007
- },
3008
- {
3009
- "epoch": 0.9212433564181027,
3010
- "grad_norm": 11.682183265686035,
3011
- "learning_rate": 1.6119316859383837e-07,
3012
- "loss": 1.2778,
3013
- "step": 4290
3014
- },
3015
- {
3016
- "epoch": 0.9233907768293338,
3017
- "grad_norm": 7.454334259033203,
3018
- "learning_rate": 1.5254958992280022e-07,
3019
- "loss": 2.2826,
3020
- "step": 4300
3021
- },
3022
- {
3023
- "epoch": 0.9255381972405647,
3024
- "grad_norm": 10.836345672607422,
3025
- "learning_rate": 1.4414059801462709e-07,
3026
- "loss": 1.9795,
3027
- "step": 4310
3028
- },
3029
- {
3030
- "epoch": 0.9276856176517958,
3031
- "grad_norm": 17.844743728637695,
3032
- "learning_rate": 1.3596659981277016e-07,
3033
- "loss": 2.6067,
3034
- "step": 4320
3035
- },
3036
- {
3037
- "epoch": 0.9298330380630268,
3038
- "grad_norm": 7.685535907745361,
3039
- "learning_rate": 1.280279908884291e-07,
3040
- "loss": 1.5424,
3041
- "step": 4330
3042
- },
3043
- {
3044
- "epoch": 0.9319804584742578,
3045
- "grad_norm": 7.696123123168945,
3046
- "learning_rate": 1.2032515542141188e-07,
3047
- "loss": 1.5663,
3048
- "step": 4340
3049
- },
3050
- {
3051
- "epoch": 0.9341278788854888,
3052
- "grad_norm": 8.067780494689941,
3053
- "learning_rate": 1.128584661815435e-07,
3054
- "loss": 1.5158,
3055
- "step": 4350
3056
- },
3057
- {
3058
- "epoch": 0.9362752992967198,
3059
- "grad_norm": 7.884578227996826,
3060
- "learning_rate": 1.0562828451062323e-07,
3061
- "loss": 2.2029,
3062
- "step": 4360
3063
- },
3064
- {
3065
- "epoch": 0.9384227197079508,
3066
- "grad_norm": 9.749954223632812,
3067
- "learning_rate": 9.86349603049419e-08,
3068
- "loss": 1.566,
3069
- "step": 4370
3070
- },
3071
- {
3072
- "epoch": 0.9405701401191818,
3073
- "grad_norm": 9.595564842224121,
3074
- "learning_rate": 9.187883199834491e-08,
3075
- "loss": 1.5778,
3076
- "step": 4380
3077
- },
3078
- {
3079
- "epoch": 0.9427175605304129,
3080
- "grad_norm": 9.646600723266602,
3081
- "learning_rate": 8.536022654585752e-08,
3082
- "loss": 1.2144,
3083
- "step": 4390
3084
- },
3085
- {
3086
- "epoch": 0.9448649809416438,
3087
- "grad_norm": 10.880244255065918,
3088
- "learning_rate": 7.907945940786033e-08,
3089
- "loss": 2.6352,
3090
- "step": 4400
3091
- },
3092
- {
3093
- "epoch": 0.9470124013528749,
3094
- "grad_norm": 18.716659545898438,
3095
- "learning_rate": 7.303683453482368e-08,
3096
- "loss": 3.4305,
3097
- "step": 4410
3098
- },
3099
- {
3100
- "epoch": 0.9491598217641058,
3101
- "grad_norm": 9.710079193115234,
3102
- "learning_rate": 6.723264435259725e-08,
3103
- "loss": 2.7579,
3104
- "step": 4420
3105
- },
3106
- {
3107
- "epoch": 0.9513072421753369,
3108
- "grad_norm": 8.077021598815918,
3109
- "learning_rate": 6.16671697482607e-08,
3110
- "loss": 1.9376,
3111
- "step": 4430
3112
- },
3113
- {
3114
- "epoch": 0.9534546625865679,
3115
- "grad_norm": 10.034996032714844,
3116
- "learning_rate": 5.634068005652804e-08,
3117
- "loss": 2.3081,
3118
- "step": 4440
3119
- },
3120
- {
3121
- "epoch": 0.9556020829977989,
3122
- "grad_norm": 10.336363792419434,
3123
- "learning_rate": 5.125343304671459e-08,
3124
- "loss": 2.3248,
3125
- "step": 4450
3126
- },
3127
- {
3128
- "epoch": 0.9577495034090299,
3129
- "grad_norm": 8.281291007995605,
3130
- "learning_rate": 4.640567491026316e-08,
3131
- "loss": 1.596,
3132
- "step": 4460
3133
- },
3134
- {
3135
- "epoch": 0.9598969238202609,
3136
- "grad_norm": 9.9288969039917,
3137
- "learning_rate": 4.179764024882793e-08,
3138
- "loss": 1.983,
3139
- "step": 4470
3140
- },
3141
- {
3142
- "epoch": 0.9620443442314919,
3143
- "grad_norm": 8.1442289352417,
3144
- "learning_rate": 3.7429552062923644e-08,
3145
- "loss": 1.4951,
3146
- "step": 4480
3147
- },
3148
- {
3149
- "epoch": 0.9641917646427229,
3150
- "grad_norm": 9.167194366455078,
3151
- "learning_rate": 3.330162174113249e-08,
3152
- "loss": 1.8853,
3153
- "step": 4490
3154
- },
3155
- {
3156
- "epoch": 0.9663391850539539,
3157
- "grad_norm": 8.423112869262695,
3158
- "learning_rate": 2.9414049049872883e-08,
3159
- "loss": 1.8813,
3160
- "step": 4500
3161
- },
3162
- {
3163
- "epoch": 0.968486605465185,
3164
- "grad_norm": 9.233404159545898,
3165
- "learning_rate": 2.5767022123734942e-08,
3166
- "loss": 1.9462,
3167
- "step": 4510
3168
- },
3169
- {
3170
- "epoch": 0.970634025876416,
3171
- "grad_norm": 18.79878044128418,
3172
- "learning_rate": 2.236071745637336e-08,
3173
- "loss": 2.2862,
3174
- "step": 4520
3175
- },
3176
- {
3177
- "epoch": 0.972781446287647,
3178
- "grad_norm": 7.818319797515869,
3179
- "learning_rate": 1.9195299891968667e-08,
3180
- "loss": 1.5587,
3181
- "step": 4530
3182
- },
3183
- {
3184
- "epoch": 0.974928866698878,
3185
- "grad_norm": 14.589919090270996,
3186
- "learning_rate": 1.627092261724583e-08,
3187
- "loss": 2.3639,
3188
- "step": 4540
3189
- },
3190
- {
3191
- "epoch": 0.977076287110109,
3192
- "grad_norm": 8.41909408569336,
3193
- "learning_rate": 1.35877271540652e-08,
3194
- "loss": 2.2179,
3195
- "step": 4550
3196
- },
3197
- {
3198
- "epoch": 0.97922370752134,
3199
- "grad_norm": 18.559860229492188,
3200
- "learning_rate": 1.1145843352572406e-08,
3201
- "loss": 2.0184,
3202
- "step": 4560
3203
- },
3204
- {
3205
- "epoch": 0.981371127932571,
3206
- "grad_norm": 11.833897590637207,
3207
- "learning_rate": 8.945389384912851e-09,
3208
- "loss": 1.9478,
3209
- "step": 4570
3210
- },
3211
- {
3212
- "epoch": 0.983518548343802,
3213
- "grad_norm": 19.6647891998291,
3214
- "learning_rate": 6.986471739513501e-09,
3215
- "loss": 1.9924,
3216
- "step": 4580
3217
- },
3218
- {
3219
- "epoch": 0.985665968755033,
3220
- "grad_norm": 7.491504669189453,
3221
- "learning_rate": 5.269185215931449e-09,
3222
- "loss": 1.2046,
3223
- "step": 4590
3224
- },
3225
- {
3226
- "epoch": 0.987813389166264,
3227
- "grad_norm": 8.722803115844727,
3228
- "learning_rate": 3.7936129202648106e-09,
3229
- "loss": 1.5883,
3230
- "step": 4600
3231
- },
3232
- {
3233
- "epoch": 0.989960809577495,
3234
- "grad_norm": 8.63818073272705,
3235
- "learning_rate": 2.5598262611298275e-09,
3236
- "loss": 1.2219,
3237
- "step": 4610
3238
- },
3239
- {
3240
- "epoch": 0.992108229988726,
3241
- "grad_norm": 9.115127563476562,
3242
- "learning_rate": 1.5678849462058554e-09,
3243
- "loss": 2.692,
3244
- "step": 4620
3245
- },
3246
- {
3247
- "epoch": 0.994255650399957,
3248
- "grad_norm": 18.569103240966797,
3249
- "learning_rate": 8.178369793482299e-10,
3250
- "loss": 2.3245,
3251
- "step": 4630
3252
- },
3253
- {
3254
- "epoch": 0.9964030708111881,
3255
- "grad_norm": 17.41649055480957,
3256
- "learning_rate": 3.097186582606826e-10,
3257
- "loss": 3.4339,
3258
- "step": 4640
3259
- },
3260
- {
3261
- "epoch": 0.9985504912224191,
3262
- "grad_norm": 7.883552551269531,
3263
- "learning_rate": 4.3554572743409463e-11,
3264
- "loss": 2.5974,
3265
- "step": 4650
3266
- },
3267
- {
3268
- "epoch": 0.9998389434691577,
3269
- "step": 4656,
3270
- "total_flos": 3.7794240074694525e+19,
3271
- "train_loss": 2.209910297516695,
3272
- "train_runtime": 29829.0342,
3273
- "train_samples_per_second": 9.991,
3274
- "train_steps_per_second": 0.156
3275
- }
3276
- ],
3277
- "logging_steps": 10,
3278
- "max_steps": 4656,
3279
- "num_input_tokens_seen": 0,
3280
- "num_train_epochs": 1,
3281
- "save_steps": 100000,
3282
- "stateful_callbacks": {
3283
- "TrainerControl": {
3284
- "args": {
3285
- "should_epoch_stop": false,
3286
- "should_evaluate": false,
3287
- "should_log": false,
3288
- "should_save": true,
3289
- "should_training_stop": true
3290
- },
3291
- "attributes": {}
3292
- }
3293
- },
3294
- "total_flos": 3.7794240074694525e+19,
3295
- "train_batch_size": 1,
3296
- "trial_name": null,
3297
- "trial_params": null
3298
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:046e63488d2ef3c054c8e281f6d7bdec5d13667b525e5e08db174b565281ce2d
3
- size 7352
 
 
 
 
QeoThinker-VGGT-Qwen25VL-7B-Vanilla/vocab.json DELETED
The diff for this file is too large to render. See raw diff