Hugging Face's logo

huihui-ai
/
Huihui-gemma-3n-E4B-it-abliterated

Image-Text-to-Text
Transformers
Safetensors
gemma3n
automatic-speech-recognition
automatic-speech-translation
audio-text-to-text
video-text-to-text
abliterated
uncensored
conversational
Model card Files Community
5
huihui-ai commited on
Commit
dbd167a
·
verified ·
1 Parent(s): 32d36f1

Add files using upload-large-folder tool

Browse files
Files changed (15) hide show
  1. .gitattributes +1 -0
  2. chat_template.jinja +49 -0
  3. config.json +189 -0
  4. generation_config.json +13 -0
  5. model-00001-of-00004.safetensors +3 -0
  6. model-00002-of-00004.safetensors +3 -0
  7. model-00003-of-00004.safetensors +3 -0
  8. model-00004-of-00004.safetensors +3 -0
  9. model.safetensors.index.json +0 -0
  10. preprocessor_config.json +51 -0
  11. processor_config.json +5 -0
  12. special_tokens_map.json +36 -0
  13. tokenizer.json +3 -0
  14. tokenizer.model +3 -0
  15. tokenizer_config.json +0 -0
.gitattributes CHANGED
@@ -34,3 +34,4 @@ saved_model/**/* 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
  bee.jpg 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
  bee.jpg filter=lfs diff=lfs merge=lfs -text
37
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
chat_template.jinja ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {{ bos_token }}
2
+ {%- if messages[0]['role'] == 'system' -%}
3
+ {%- if messages[0]['content'] is string -%}
4
+ {%- set first_user_prefix = messages[0]['content'] + '
5
+
6
+ ' -%}
7
+ {%- else -%}
8
+ {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
9
+
10
+ ' -%}
11
+ {%- endif -%}
12
+ {%- set loop_messages = messages[1:] -%}
13
+ {%- else -%}
14
+ {%- set first_user_prefix = "" -%}
15
+ {%- set loop_messages = messages -%}
16
+ {%- endif -%}
17
+ {%- for message in loop_messages -%}
18
+ {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
19
+ {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
20
+ {%- endif -%}
21
+ {%- if (message['role'] == 'assistant') -%}
22
+ {%- set role = "model" -%}
23
+ {%- else -%}
24
+ {%- set role = message['role'] -%}
25
+ {%- endif -%}
26
+ {{ '<start_of_turn>' + role + '
27
+ ' + (first_user_prefix if loop.first else "") }}
28
+ {%- if message['content'] is string -%}
29
+ {{ message['content'] | trim }}
30
+ {%- elif message['content'] is iterable -%}
31
+ {%- for item in message['content'] -%}
32
+ {%- if item['type'] == 'audio' -%}
33
+ {{ '<audio_soft_token>' }}
34
+ {%- elif item['type'] == 'image' -%}
35
+ {{ '<image_soft_token>' }}
36
+ {%- elif item['type'] == 'text' -%}
37
+ {{ item['text'] | trim }}
38
+ {%- endif -%}
39
+ {%- endfor -%}
40
+ {%- else -%}
41
+ {{ raise_exception("Invalid content type") }}
42
+ {%- endif -%}
43
+ {{ '<end_of_turn>
44
+ ' }}
45
+ {%- endfor -%}
46
+ {%- if add_generation_prompt -%}
47
+ {{'<start_of_turn>model
48
+ '}}
49
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,189 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3nForConditionalGeneration"
4
+ ],
5
+ "audio_config": {
6
+ "conf_attention_chunk_size": 12,
7
+ "conf_attention_context_left": 13,
8
+ "conf_attention_context_right": 0,
9
+ "conf_attention_logit_cap": 50.0,
10
+ "conf_conv_kernel_size": 5,
11
+ "conf_num_attention_heads": 8,
12
+ "conf_num_hidden_layers": 12,
13
+ "conf_positional_bias_size": 256,
14
+ "conf_reduction_factor": 4,
15
+ "conf_residual_weight": 0.5,
16
+ "gradient_clipping": 10000000000.0,
17
+ "hidden_size": 1536,
18
+ "input_feat_size": 128,
19
+ "model_type": "gemma3n_audio",
20
+ "rms_norm_eps": 1e-06,
21
+ "sscp_conv_channel_size": [
22
+ 128,
23
+ 32
24
+ ],
25
+ "sscp_conv_eps": 0.001,
26
+ "sscp_conv_kernel_size": [
27
+ [
28
+ 3,
29
+ 3
30
+ ],
31
+ [
32
+ 3,
33
+ 3
34
+ ]
35
+ ],
36
+ "sscp_conv_stride_size": [
37
+ [
38
+ 2,
39
+ 2
40
+ ],
41
+ [
42
+ 2,
43
+ 2
44
+ ]
45
+ ],
46
+ "torch_dtype": "bfloat16",
47
+ "vocab_offset": 262272,
48
+ "vocab_size": 128
49
+ },
50
+ "audio_soft_tokens_per_image": 188,
51
+ "audio_token_id": 262273,
52
+ "boa_token_id": 256000,
53
+ "boi_token_id": 255999,
54
+ "eoa_token_id": 262272,
55
+ "eoi_token_id": 262144,
56
+ "eos_token_id": [
57
+ 1,
58
+ 106
59
+ ],
60
+ "image_token_id": 262145,
61
+ "initializer_range": 0.02,
62
+ "model_type": "gemma3n",
63
+ "text_config": {
64
+ "activation_sparsity_pattern": [
65
+ 0.95,
66
+ 0.95,
67
+ 0.95,
68
+ 0.95,
69
+ 0.95,
70
+ 0.95,
71
+ 0.95,
72
+ 0.95,
73
+ 0.95,
74
+ 0.95,
75
+ 0.0,
76
+ 0.0,
77
+ 0.0,
78
+ 0.0,
79
+ 0.0,
80
+ 0.0,
81
+ 0.0,
82
+ 0.0,
83
+ 0.0,
84
+ 0.0,
85
+ 0.0,
86
+ 0.0,
87
+ 0.0,
88
+ 0.0,
89
+ 0.0,
90
+ 0.0,
91
+ 0.0,
92
+ 0.0,
93
+ 0.0,
94
+ 0.0,
95
+ 0.0,
96
+ 0.0,
97
+ 0.0,
98
+ 0.0,
99
+ 0.0
100
+ ],
101
+ "altup_active_idx": 0,
102
+ "altup_coef_clip": 120.0,
103
+ "altup_correct_scale": true,
104
+ "altup_lr_multiplier": 1.0,
105
+ "altup_num_inputs": 4,
106
+ "attention_bias": false,
107
+ "attention_dropout": 0.0,
108
+ "final_logit_softcapping": 30.0,
109
+ "head_dim": 256,
110
+ "hidden_activation": "gelu_pytorch_tanh",
111
+ "hidden_size": 2048,
112
+ "hidden_size_per_layer_input": 256,
113
+ "initializer_range": 0.02,
114
+ "intermediate_size": 16384,
115
+ "laurel_rank": 64,
116
+ "layer_types": [
117
+ "sliding_attention",
118
+ "sliding_attention",
119
+ "sliding_attention",
120
+ "sliding_attention",
121
+ "full_attention",
122
+ "sliding_attention",
123
+ "sliding_attention",
124
+ "sliding_attention",
125
+ "sliding_attention",
126
+ "full_attention",
127
+ "sliding_attention",
128
+ "sliding_attention",
129
+ "sliding_attention",
130
+ "sliding_attention",
131
+ "full_attention",
132
+ "sliding_attention",
133
+ "sliding_attention",
134
+ "sliding_attention",
135
+ "sliding_attention",
136
+ "full_attention",
137
+ "sliding_attention",
138
+ "sliding_attention",
139
+ "sliding_attention",
140
+ "sliding_attention",
141
+ "full_attention",
142
+ "sliding_attention",
143
+ "sliding_attention",
144
+ "sliding_attention",
145
+ "sliding_attention",
146
+ "full_attention",
147
+ "sliding_attention",
148
+ "sliding_attention",
149
+ "sliding_attention",
150
+ "sliding_attention",
151
+ "full_attention"
152
+ ],
153
+ "max_position_embeddings": 32768,
154
+ "model_type": "gemma3n_text",
155
+ "num_attention_heads": 8,
156
+ "num_hidden_layers": 35,
157
+ "num_key_value_heads": 2,
158
+ "num_kv_shared_layers": 15,
159
+ "query_pre_attn_scalar": 256,
160
+ "rms_norm_eps": 1e-06,
161
+ "rope_local_base_freq": 10000.0,
162
+ "rope_scaling": null,
163
+ "rope_theta": 1000000.0,
164
+ "sliding_window": 512,
165
+ "torch_dtype": "bfloat16",
166
+ "use_cache": true,
167
+ "vocab_size": 262400,
168
+ "vocab_size_per_layer_input": 262144
169
+ },
170
+ "torch_dtype": "bfloat16",
171
+ "transformers_version": "4.53.0.dev0",
172
+ "vision_config": {
173
+ "architecture": "mobilenetv5_300m_enc",
174
+ "do_pooling": true,
175
+ "hidden_size": 2048,
176
+ "initializer_range": 0.02,
177
+ "label_names": [
178
+ "LABEL_0",
179
+ "LABEL_1"
180
+ ],
181
+ "model_type": "gemma3n_vision",
182
+ "num_classes": 2,
183
+ "rms_norm_eps": 1e-06,
184
+ "torch_dtype": "bfloat16",
185
+ "vocab_offset": 262144,
186
+ "vocab_size": 128
187
+ },
188
+ "vision_soft_tokens_per_image": 256
189
+ }
generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1,
7
+ 106
8
+ ],
9
+ "pad_token_id": 0,
10
+ "top_k": 64,
11
+ "top_p": 0.95,
12
+ "transformers_version": "4.53.0.dev0"
13
+ }
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d3abb4293f4ab6a3542bdfb15204cf6784a434db7d48a0004725ca1951c45d7
3
+ size 4967933712
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fce45808b9e2047bf20e97e6501b7d9ea09090e349ca5cf1b32f693b11370ba
3
+ size 4569247128
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5597d11cd53e6830d026588b38c7491ebd104820fde3fd7d4a0b743591f352e
3
+ size 4995580968
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b215d940b1442e27d2a33070fd871c5e08973599517ae35226ae02e058afa581
3
+ size 1167419608
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": null,
3
+ "data_format": "channels_first",
4
+ "default_to_square": false,
5
+ "device": null,
6
+ "dither": 0.0,
7
+ "do_center_crop": null,
8
+ "do_convert_rgb": null,
9
+ "do_normalize": false,
10
+ "do_rescale": true,
11
+ "do_resize": true,
12
+ "feature_extractor_type": "Gemma3nAudioFeatureExtractor",
13
+ "feature_size": 128,
14
+ "fft_length": 1024,
15
+ "fft_overdrive": true,
16
+ "frame_length": 512,
17
+ "hop_length": 160,
18
+ "image_mean": [
19
+ 0.5,
20
+ 0.5,
21
+ 0.5
22
+ ],
23
+ "image_processor_type": "SiglipImageProcessorFast",
24
+ "image_seq_length": 256,
25
+ "image_std": [
26
+ 0.5,
27
+ 0.5,
28
+ 0.5
29
+ ],
30
+ "input_data_format": null,
31
+ "input_scale_factor": 1.0,
32
+ "max_frequency": 7600.0,
33
+ "mel_floor": 1e-05,
34
+ "min_frequency": 125.0,
35
+ "padding_side": "right",
36
+ "padding_value": 0.0,
37
+ "per_bin_mean": null,
38
+ "per_bin_stddev": null,
39
+ "preemphasis": 0.97,
40
+ "preemphasis_htk_flavor": true,
41
+ "processor_class": "Gemma3nProcessor",
42
+ "resample": 2,
43
+ "rescale_factor": 0.00392156862745098,
44
+ "return_attention_mask": false,
45
+ "return_tensors": null,
46
+ "sampling_rate": 16000,
47
+ "size": {
48
+ "height": 768,
49
+ "width": 768
50
+ }
51
+ }
processor_config.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "audio_seq_length": 188,
3
+ "image_seq_length": 256,
4
+ "processor_class": "Gemma3nProcessor"
5
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "audio_token": "<audio_soft_token>",
3
+ "boa_token": "<start_of_audio>",
4
+ "boi_token": "<start_of_image>",
5
+ "bos_token": {
6
+ "content": "<bos>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "eoa_token": "<end_of_audio>",
13
+ "eoi_token": "<end_of_image>",
14
+ "eos_token": {
15
+ "content": "<eos>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "image_token": "<image_soft_token>",
22
+ "pad_token": {
23
+ "content": "<pad>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false
28
+ },
29
+ "unk_token": {
30
+ "content": "<unk>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false
35
+ }
36
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4c19736bf24d1c6805cf49340e31bd02c70fb7857a2cb31065c90c2b5719c4e
3
+ size 33442559
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea5f0cc48abfbfc04d14562270a32e02149a3e7035f368cc5a462786f4a59961
3
+ size 4696020
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff