CCATresearch commited on
Commit
1f5d13f
·
verified ·
1 Parent(s): 996f97d

Upload 7 files

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
chat_template.jinja ADDED
@@ -0,0 +1 @@
 
 
1
+ {{ bos_token }}{%- for message in messages %}{%- if message.role == 'system' %}{{- '<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>' }}{%- if message.content is string %}{{- message.content }}{%- else %}{%- for content in message.content %}{%- if 'text' in content %}{{- content.text }}{%- endif %}{%- endfor %}{%- endif %}{{- '<|END_OF_TURN_TOKEN|>' }}{%- elif message.role == 'user' %}{{- '<|START_OF_TURN_TOKEN|><|USER_TOKEN|>' }}{%- if message.content is string %}{{- message.content }}{%- else %}{%- for content in message.content %}{%- if content.type == 'image' or 'image' in content or 'image_url' in content %}<|VISION_START|><|IMAGE_PAD|><|VISION_END|>{%- elif content.type == 'video' or 'video' in content %}<|VISION_START|><|VIDEO_PAD|><|VISION_END|>{%- elif 'text' in content %}{{- content.text }}{%- endif %}{%- endfor %}{%- endif %}{{- '<|END_OF_TURN_TOKEN|>' }}{%- elif message.role == 'assistant' or message.role == 'chatbot' %}{{- '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' }}{%- if message.content is string %}{{- message.content }}{%- else %}{%- for content in message.content %}{%- if 'text' in content %}{{- content.text }}{%- endif %}{%- endfor %}{%- endif %}{{- '<|END_OF_TURN_TOKEN|>' }}{%- endif %}{%- endfor %}{%- if add_generation_prompt %}{{- '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' }}{%- endif %}
config.json ADDED
@@ -0,0 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CohereCompassForConditionalGeneration"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "eos_token_id": 255001,
7
+ "fusion_config": {
8
+ "patch_embeddings": true
9
+ },
10
+ "image_token_id": 255031,
11
+ "model_type": "cohere_compass",
12
+ "pad_token_id": 0,
13
+ "text_config": {
14
+ "attention_bias": false,
15
+ "attention_dropout": 0.0,
16
+ "bos_token_id": 2,
17
+ "dtype": "bfloat16",
18
+ "eos_token_id": 255001,
19
+ "head_dim": 128,
20
+ "hidden_act": "silu",
21
+ "hidden_size": 2048,
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 6144,
24
+ "layer_norm_eps": 1e-05,
25
+ "layer_types": [
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "full_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "full_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "full_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "full_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "sliding_attention",
49
+ "full_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "full_attention"
54
+ ],
55
+ "logit_scale": 0.25,
56
+ "max_position_embeddings": 500000,
57
+ "model_type": "cohere_compass_text",
58
+ "num_attention_heads": 16,
59
+ "num_hidden_layers": 28,
60
+ "num_key_value_heads": 8,
61
+ "pad_token_id": 0,
62
+ "pooling": null,
63
+ "rope_parameters": {
64
+ "full_attention": null,
65
+ "rope_theta": 10000.0,
66
+ "rope_type": "default",
67
+ "sliding_attention": {
68
+ "mrope_interleaved": true,
69
+ "mrope_section": [
70
+ 24,
71
+ 20,
72
+ 20
73
+ ],
74
+ "rope_theta": 50000,
75
+ "rope_type": "default"
76
+ }
77
+ },
78
+ "score_shift_a": null,
79
+ "score_shift_b": null,
80
+ "sliding_window": 4096,
81
+ "tie_word_embeddings": true,
82
+ "use_cache": true,
83
+ "vocab_size": 262144
84
+ },
85
+ "tie_word_embeddings": true,
86
+ "transformers_version": "5.16.0.dev0",
87
+ "video_token_id": 255032,
88
+ "vision_config": {
89
+ "deepstack_visual_indexes": [
90
+ 8,
91
+ 16,
92
+ 24
93
+ ],
94
+ "depth": 27,
95
+ "dtype": "bfloat16",
96
+ "hidden_act": "gelu_pytorch_tanh",
97
+ "hidden_size": 1152,
98
+ "in_channels": 3,
99
+ "initializer_range": 0.02,
100
+ "intermediate_size": 4304,
101
+ "model_type": "cohere_compass_vision",
102
+ "num_heads": 16,
103
+ "num_position_embeddings": 2304,
104
+ "out_hidden_size": 2048,
105
+ "patch_size": 16,
106
+ "spatial_merge_size": 2,
107
+ "temporal_patch_size": 2
108
+ },
109
+ "vision_end_token_id": 255029,
110
+ "vision_start_token_id": 255028
111
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 255001
6
+ ],
7
+ "pad_token_id": 0,
8
+ "temperature": 0.7,
9
+ "top_k": 20,
10
+ "top_p": 0.8,
11
+ "transformers_version": "5.16.0.dev0"
12
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfd4ccbea1344f417120d52f9efd950afac605ef2c7df4ee52324a72fb437e59
3
+ size 4969765248
processor_config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_processor": {
3
+ "data_format": "channels_first",
4
+ "default_to_square": false,
5
+ "do_convert_rgb": true,
6
+ "do_normalize": true,
7
+ "do_rescale": true,
8
+ "do_resize": true,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_processor_type": "CohereCompassImageProcessor",
15
+ "image_std": [
16
+ 0.5,
17
+ 0.5,
18
+ 0.5
19
+ ],
20
+ "merge_size": 2,
21
+ "patch_size": 16,
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "longest_edge": 16777216,
26
+ "shortest_edge": 65536
27
+ },
28
+ "temporal_patch_size": 2
29
+ },
30
+ "processor_class": "CohereCompassProcessor",
31
+ "video_processor": {
32
+ "do_convert_rgb": true,
33
+ "do_normalize": true,
34
+ "do_rescale": true,
35
+ "do_resize": true,
36
+ "do_sample_frames": true,
37
+ "fps": 2,
38
+ "image_mean": [
39
+ 0.5,
40
+ 0.5,
41
+ 0.5
42
+ ],
43
+ "image_std": [
44
+ 0.5,
45
+ 0.5,
46
+ 0.5
47
+ ],
48
+ "max_frames": 768,
49
+ "merge_size": 2,
50
+ "min_frames": 4,
51
+ "patch_size": 16,
52
+ "resample": 3,
53
+ "rescale_factor": 0.00392156862745098,
54
+ "return_metadata": false,
55
+ "size": {
56
+ "longest_edge": 25165824,
57
+ "shortest_edge": 4096
58
+ },
59
+ "temporal_patch_size": 2,
60
+ "video_processor_type": "CohereCompassVideoProcessor"
61
+ }
62
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4594953eca27aa1122252e8a61a91a5b4941c92aba78d3811c8fe9b1b95d7f68
3
+ size 19550475
tokenizer_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<BOS_TOKEN>",
5
+ "clean_up_tokenization_spaces": false,
6
+ "cls_token": "<CLS>",
7
+ "eos_token": "<|END_OF_TURN_TOKEN|>",
8
+ "errors": "replace",
9
+ "image_token": "<|IMAGE_PAD|>",
10
+ "is_local": true,
11
+ "legacy": true,
12
+ "local_files_only": false,
13
+ "mask_token": "<MASK_TOKEN>",
14
+ "max_pixels": 3868706,
15
+ "min_pixels": 16384,
16
+ "model_max_length": 1000000000000000019884624838656,
17
+ "model_specific_special_tokens": {
18
+ "image_token": "<|IMAGE_PAD|>",
19
+ "video_token": "<|VIDEO_PAD|>",
20
+ "vision_end_token": "<|VISION_END|>",
21
+ "vision_start_token": "<|VISION_START|>"
22
+ },
23
+ "pad_token": "<PAD>",
24
+ "padding_side": "right",
25
+ "processor_class": "CohereCompassProcessor",
26
+ "sep_token": "<SEP>",
27
+ "sp_model_kwargs": {},
28
+ "spaces_between_special_tokens": false,
29
+ "tokenizer_class": "CohereTokenizer",
30
+ "unk_token": "<UNK>",
31
+ "use_default_system_prompt": false,
32
+ "video_token": "<|VIDEO_PAD|>",
33
+ "vision_end_token": "<|VISION_END|>",
34
+ "vision_start_token": "<|VISION_START|>"
35
+ }