Hugging Face's logo

unsloth
/
gemma-3n-E2B-it-unsloth-bnb-4bit

Image-Text-to-Text
Transformers
Safetensors
English
gemma3n
gemma3
unsloth
gemma
google
conversational
4-bit precision
bitsandbytes
Model card Files
xet
Community
2
danielhanchen commited on
Commit
0bbf790
·
verified ·
1 Parent(s): 54c7439

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +38 -5
  2. generation_config.json +1 -1
  3. model-00001-of-00003.safetensors +2 -2
  4. model-00002-of-00003.safetensors +2 -2
  5. model-00003-of-00003.safetensors +2 -2
  6. model.safetensors.index.json +0 -0
  7. preprocessor_config.json +1 -0
config.json CHANGED
@@ -23,6 +23,7 @@
23
  32
24
  ],
25
  "sscp_conv_eps": 0.001,
 
26
  "sscp_conv_kernel_size": [
27
  [
28
  3,
@@ -80,12 +81,12 @@
80
  "modality_router",
81
  "embedding_projection",
82
  "input_proj_linear",
83
- "laurel",
84
- "per_layer_input_gate",
85
  "per_layer_projection",
86
  "audio_tower",
87
  "embed_vision",
88
- "embed_audio"
 
 
89
  ],
90
  "llm_int8_threshold": 6.0,
91
  "load_in_4bit": true,
@@ -138,7 +139,38 @@
138
  "hidden_size": 2048,
139
  "hidden_size_per_layer_input": 256,
140
  "initializer_range": 0.02,
141
- "intermediate_size": 8192,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
142
  "laurel_rank": 64,
143
  "layer_types": [
144
  "sliding_attention",
@@ -190,7 +222,7 @@
190
  "vocab_size_per_layer_input": 262144
191
  },
192
  "torch_dtype": "bfloat16",
193
- "transformers_version": "4.53.0.dev0",
194
  "unsloth_fixed": true,
195
  "vision_config": {
196
  "architecture": "mobilenetv5_300m_enc",
@@ -201,6 +233,7 @@
201
  "LABEL_0",
202
  "LABEL_1"
203
  ],
 
204
  "model_type": "gemma3n_vision",
205
  "num_classes": 2,
206
  "rms_norm_eps": 1e-06,
 
23
  32
24
  ],
25
  "sscp_conv_eps": 0.001,
26
+ "sscp_conv_group_norm_eps": 0.001,
27
  "sscp_conv_kernel_size": [
28
  [
29
  3,
 
81
  "modality_router",
82
  "embedding_projection",
83
  "input_proj_linear",
 
 
84
  "per_layer_projection",
85
  "audio_tower",
86
  "embed_vision",
87
+ "embed_audio",
88
+ "altup_projections",
89
+ "altup_unembed_projections"
90
  ],
91
  "llm_int8_threshold": 6.0,
92
  "load_in_4bit": true,
 
139
  "hidden_size": 2048,
140
  "hidden_size_per_layer_input": 256,
141
  "initializer_range": 0.02,
142
+ "intermediate_size": [
143
+ 8192,
144
+ 8192,
145
+ 8192,
146
+ 8192,
147
+ 8192,
148
+ 8192,
149
+ 8192,
150
+ 8192,
151
+ 8192,
152
+ 8192,
153
+ 8192,
154
+ 8192,
155
+ 8192,
156
+ 8192,
157
+ 8192,
158
+ 8192,
159
+ 8192,
160
+ 8192,
161
+ 8192,
162
+ 8192,
163
+ 8192,
164
+ 8192,
165
+ 8192,
166
+ 8192,
167
+ 8192,
168
+ 8192,
169
+ 8192,
170
+ 8192,
171
+ 8192,
172
+ 8192
173
+ ],
174
  "laurel_rank": 64,
175
  "layer_types": [
176
  "sliding_attention",
 
222
  "vocab_size_per_layer_input": 262144
223
  },
224
  "torch_dtype": "bfloat16",
225
+ "transformers_version": "4.53.0",
226
  "unsloth_fixed": true,
227
  "vision_config": {
228
  "architecture": "mobilenetv5_300m_enc",
 
233
  "LABEL_0",
234
  "LABEL_1"
235
  ],
236
+ "model_args": null,
237
  "model_type": "gemma3n_vision",
238
  "num_classes": 2,
239
  "rms_norm_eps": 1e-06,
generation_config.json CHANGED
@@ -9,5 +9,5 @@
9
  "pad_token_id": 0,
10
  "top_k": 64,
11
  "top_p": 0.95,
12
- "transformers_version": "4.53.0.dev0"
13
  }
 
9
  "pad_token_id": 0,
10
  "top_k": 64,
11
  "top_p": 0.95,
12
+ "transformers_version": "4.53.0"
13
  }
model-00001-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:16d40687c46a2d0b8064c1c0d3281e04bd1a5b7feb9c2b40849dbf07dceb3dea
3
- size 2684992260
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fdc029a2bc6493b52d91d8d649742c648630b703924cada9eeb53601447434e
3
+ size 2650152631
model-00002-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ce5b74cd7f804d2ba4278a1498a02e435ca5e82baa6959d198225282952caa2e
3
- size 4999684414
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a68539f95af3848759c88e426d9bf91714b9be2917ec2d67d09828114fd5e433
3
+ size 4994547572
model-00003-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:36b6e7f527f75e3e32a7a977c474cbe3a1f32b8df284129ccc7efe48990c5577
3
- size 426305256
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd9207d856bcea59ac0979141503f2de5c8ad2f20a3b7feabdf0fb613b1c6d9c
3
+ size 468779960
model.safetensors.index.json CHANGED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json CHANGED
@@ -3,6 +3,7 @@
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,
 
3
  "data_format": "channels_first",
4
  "default_to_square": false,
5
  "device": null,
6
+ "disable_grouping": null,
7
  "dither": 0.0,
8
  "do_center_crop": null,
9
  "do_convert_rgb": null,