danielhanchen commited on
Commit
7d8046d
·
verified ·
1 Parent(s): 4c9396f

Upload folder using huggingface_hub

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
config.json ADDED
@@ -0,0 +1,179 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "bos_token_id": 2,
55
+ "eoa_token_id": 262272,
56
+ "eoi_token_id": 262144,
57
+ "eos_token_id": 1,
58
+ "image_token_id": 262145,
59
+ "initializer_range": 0.02,
60
+ "model_type": "gemma3n",
61
+ "pad_token_id": 0,
62
+ "text_config": {
63
+ "activation_sparsity_pattern": [
64
+ 0.95,
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
+ 0.0,
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
+ ],
95
+ "altup_active_idx": 0,
96
+ "altup_coef_clip": 120.0,
97
+ "altup_correct_scale": true,
98
+ "altup_lr_multiplier": 1.0,
99
+ "altup_num_inputs": 4,
100
+ "attention_bias": false,
101
+ "attention_dropout": 0.0,
102
+ "final_logit_softcapping": 30.0,
103
+ "head_dim": 256,
104
+ "hidden_activation": "gelu_pytorch_tanh",
105
+ "hidden_size": 2048,
106
+ "hidden_size_per_layer_input": 256,
107
+ "initializer_range": 0.02,
108
+ "intermediate_size": 8192,
109
+ "laurel_rank": 64,
110
+ "layer_types": [
111
+ "sliding_attention",
112
+ "sliding_attention",
113
+ "sliding_attention",
114
+ "sliding_attention",
115
+ "full_attention",
116
+ "sliding_attention",
117
+ "sliding_attention",
118
+ "sliding_attention",
119
+ "sliding_attention",
120
+ "full_attention",
121
+ "sliding_attention",
122
+ "sliding_attention",
123
+ "sliding_attention",
124
+ "sliding_attention",
125
+ "full_attention",
126
+ "sliding_attention",
127
+ "sliding_attention",
128
+ "sliding_attention",
129
+ "sliding_attention",
130
+ "full_attention",
131
+ "sliding_attention",
132
+ "sliding_attention",
133
+ "sliding_attention",
134
+ "sliding_attention",
135
+ "full_attention",
136
+ "sliding_attention",
137
+ "sliding_attention",
138
+ "sliding_attention",
139
+ "sliding_attention",
140
+ "full_attention"
141
+ ],
142
+ "max_position_embeddings": 32768,
143
+ "model_type": "gemma3n_text",
144
+ "num_attention_heads": 8,
145
+ "num_hidden_layers": 30,
146
+ "num_key_value_heads": 2,
147
+ "num_kv_shared_layers": 10,
148
+ "query_pre_attn_scalar": 256,
149
+ "rms_norm_eps": 1e-06,
150
+ "rope_local_base_freq": 10000.0,
151
+ "rope_scaling": null,
152
+ "rope_theta": 1000000.0,
153
+ "sliding_window": 512,
154
+ "torch_dtype": "bfloat16",
155
+ "use_cache": true,
156
+ "vocab_size": 262400,
157
+ "vocab_size_per_layer_input": 262144
158
+ },
159
+ "torch_dtype": "bfloat16",
160
+ "transformers_version": "4.53.0.dev0",
161
+ "unsloth_fixed": true,
162
+ "vision_config": {
163
+ "architecture": "mobilenetv5_300m_enc",
164
+ "do_pooling": true,
165
+ "hidden_size": 2048,
166
+ "initializer_range": 0.02,
167
+ "label_names": [
168
+ "LABEL_0",
169
+ "LABEL_1"
170
+ ],
171
+ "model_type": "gemma3n_vision",
172
+ "num_classes": 2,
173
+ "rms_norm_eps": 1e-06,
174
+ "torch_dtype": "bfloat16",
175
+ "vocab_offset": 262144,
176
+ "vocab_size": 128
177
+ },
178
+ "vision_soft_tokens_per_image": 256
179
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": 1,
6
+ "pad_token_id": 0,
7
+ "top_k": 64,
8
+ "top_p": 0.95,
9
+ "transformers_version": "4.53.0.dev0"
10
+ }
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:681e3230979df3dcf723631523de45d7bdd69be8af6e1c348c2ceeffe17595f9
3
+ size 3077103576
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0c8bebed349589e844bb313978c3423261fd9e64cb5a811993f3b9657122db7
3
+ size 4981242176
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a186f3ca577f3eef31f3ce9eea6b8c3908c69310ed2e5e9424beff36201afbee
3
+ size 2820739840
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_size": 128,
13
+ "fft_length": 1024,
14
+ "fft_overdrive": true,
15
+ "frame_length": 512,
16
+ "hop_length": 160,
17
+ "image_mean": [
18
+ 0.5,
19
+ 0.5,
20
+ 0.5
21
+ ],
22
+ "image_processor_type": "SiglipImageProcessorFast",
23
+ "image_seq_length": 256,
24
+ "image_std": [
25
+ 0.5,
26
+ 0.5,
27
+ 0.5
28
+ ],
29
+ "input_data_format": null,
30
+ "input_scale_factor": 1.0,
31
+ "max_frequency": 7600.0,
32
+ "mel_floor": 1e-05,
33
+ "min_frequency": 125.0,
34
+ "padding_side": "left",
35
+ "padding_value": 0.0,
36
+ "per_bin_mean": null,
37
+ "per_bin_stddev": null,
38
+ "preemphasis": 0.97,
39
+ "preemphasis_htk_flavor": true,
40
+ "processor_class": "Gemma3nProcessor",
41
+ "resample": 2,
42
+ "rescale_factor": 0.00392156862745098,
43
+ "return_attention_mask": false,
44
+ "return_tensors": null,
45
+ "sampling_rate": 16000,
46
+ "size": {
47
+ "height": 768,
48
+ "width": 768
49
+ }
50
+ }
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:b6c35ee648c07754b44cd9e371c75d4caa05c4504910b7ad29b1847ee9d8ba5d
3
+ size 33442553
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