Hugging Face's logo

saddam213
/
TextEncoder

Safetensors
Model card Files
xet
 Community
saddam213 commited on
Commit
13bb701
·
verified ·
1 Parent(s): 44dfb27

Upload 11 files

Browse files
Files changed (12) hide show
  1. .gitattributes +1 -0
  2. T5-Gemma-2-4B/config.json +252 -0
  3. T5-Gemma-2-4B/generation_config.json +10 -0
  4. T5-Gemma-2-4B/model-00001-of-00004.safetensors +3 -0
  5. T5-Gemma-2-4B/model-00002-of-00004.safetensors +3 -0
  6. T5-Gemma-2-4B/model-00003-of-00004.safetensors +3 -0
  7. T5-Gemma-2-4B/model-00004-of-00004.safetensors +3 -0
  8. T5-Gemma-2-4B/model.safetensors.index.json +0 -0
  9. T5-Gemma-2-4B/processor_config.json +33 -0
  10. T5-Gemma-2-4B/special_tokens_map.json +33 -0
  11. T5-Gemma-2-4B/tokenizer.json +3 -0
  12. T5-Gemma-2-4B/tokenizer_config.json +28 -0
.gitattributes CHANGED
@@ -44,3 +44,4 @@ Qwen-3-06B/tokenizer.json filter=lfs diff=lfs merge=lfs -text
44
  Ministral-3-3B-IT/tekken.json filter=lfs diff=lfs merge=lfs -text
45
  Ministral-3-3B-IT/tokenizer.json filter=lfs diff=lfs merge=lfs -text
46
  UMT5-Base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
44
  Ministral-3-3B-IT/tekken.json filter=lfs diff=lfs merge=lfs -text
45
  Ministral-3-3B-IT/tokenizer.json filter=lfs diff=lfs merge=lfs -text
46
  UMT5-Base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
47
+ T5-Gemma-2-4B/tokenizer.json filter=lfs diff=lfs merge=lfs -text
T5-Gemma-2-4B/config.json ADDED
@@ -0,0 +1,252 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5Gemma2ForConditionalGeneration"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 2,
7
+ "classifier_dropout_rate": 0.0,
8
+ "decoder": {
9
+ "_sliding_window_pattern": 6,
10
+ "attention_bias": false,
11
+ "attention_dropout": 0.0,
12
+ "attn_logit_softcapping": null,
13
+ "dropout_rate": 0.0,
14
+ "dtype": "bfloat16",
15
+ "final_logit_softcapping": null,
16
+ "head_dim": 256,
17
+ "hidden_activation": "gelu_pytorch_tanh",
18
+ "hidden_size": 2560,
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 10240,
21
+ "layer_types": [
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "full_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "full_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "full_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "full_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "sliding_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "full_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "sliding_attention",
55
+ "sliding_attention"
56
+ ],
57
+ "max_position_embeddings": 131072,
58
+ "model_type": "t5gemma2_decoder",
59
+ "num_attention_heads": 8,
60
+ "num_hidden_layers": 34,
61
+ "num_key_value_heads": 4,
62
+ "query_pre_attn_scalar": 256,
63
+ "rms_norm_eps": 1e-06,
64
+ "rope_parameters": {
65
+ "full_attention": {
66
+ "factor": 8.0,
67
+ "rope_theta": 1000000,
68
+ "rope_type": "linear"
69
+ },
70
+ "sliding_attention": {
71
+ "rope_theta": 10000,
72
+ "rope_type": "default"
73
+ }
74
+ },
75
+ "sliding_window": 1024,
76
+ "use_bidirectional_attention": false,
77
+ "use_cache": true,
78
+ "vocab_size": 262144
79
+ },
80
+ "dropout_rate": 0.0,
81
+ "dtype": "bfloat16",
82
+ "encoder": {
83
+ "attention_dropout": 0.0,
84
+ "boi_token_index": 255999,
85
+ "dropout_rate": 0.0,
86
+ "dtype": "bfloat16",
87
+ "eoi_token_index": 256000,
88
+ "image_token_index": 256001,
89
+ "initializer_range": 0.02,
90
+ "mm_tokens_per_image": 256,
91
+ "model_type": "t5gemma2_encoder",
92
+ "text_config": {
93
+ "_name_or_path": "",
94
+ "_sliding_window_pattern": 6,
95
+ "add_cross_attention": false,
96
+ "architectures": null,
97
+ "attention_bias": false,
98
+ "attention_dropout": 0.0,
99
+ "attn_logit_softcapping": null,
100
+ "bos_token_id": 2,
101
+ "chunk_size_feed_forward": 0,
102
+ "cross_attention_hidden_size": null,
103
+ "decoder_start_token_id": null,
104
+ "dropout_rate": 0.0,
105
+ "dtype": "bfloat16",
106
+ "eos_token_id": 1,
107
+ "final_logit_softcapping": null,
108
+ "finetuning_task": null,
109
+ "head_dim": 256,
110
+ "hidden_activation": "gelu_pytorch_tanh",
111
+ "hidden_size": 2560,
112
+ "id2label": {
113
+ "0": "LABEL_0",
114
+ "1": "LABEL_1"
115
+ },
116
+ "initializer_range": 0.02,
117
+ "intermediate_size": 10240,
118
+ "is_decoder": false,
119
+ "is_encoder_decoder": false,
120
+ "label2id": {
121
+ "LABEL_0": 0,
122
+ "LABEL_1": 1
123
+ },
124
+ "layer_types": [
125
+ "sliding_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
+ "sliding_attention",
136
+ "full_attention",
137
+ "sliding_attention",
138
+ "sliding_attention",
139
+ "sliding_attention",
140
+ "sliding_attention",
141
+ "sliding_attention",
142
+ "full_attention",
143
+ "sliding_attention",
144
+ "sliding_attention",
145
+ "sliding_attention",
146
+ "sliding_attention",
147
+ "sliding_attention",
148
+ "full_attention",
149
+ "sliding_attention",
150
+ "sliding_attention",
151
+ "sliding_attention",
152
+ "sliding_attention",
153
+ "sliding_attention",
154
+ "full_attention",
155
+ "sliding_attention",
156
+ "sliding_attention",
157
+ "sliding_attention",
158
+ "sliding_attention"
159
+ ],
160
+ "max_position_embeddings": 131072,
161
+ "model_type": "t5gemma2_text",
162
+ "num_attention_heads": 8,
163
+ "num_hidden_layers": 34,
164
+ "num_key_value_heads": 4,
165
+ "output_attentions": false,
166
+ "output_hidden_states": false,
167
+ "pad_token_id": 0,
168
+ "prefix": null,
169
+ "problem_type": null,
170
+ "query_pre_attn_scalar": 256,
171
+ "return_dict": true,
172
+ "rms_norm_eps": 1e-06,
173
+ "rope_parameters": {
174
+ "full_attention": {
175
+ "factor": 8.0,
176
+ "rope_theta": 1000000,
177
+ "rope_type": "linear"
178
+ },
179
+ "sliding_attention": {
180
+ "rope_theta": 10000,
181
+ "rope_type": "default"
182
+ }
183
+ },
184
+ "sep_token_id": null,
185
+ "sliding_window": 1024,
186
+ "task_specific_params": null,
187
+ "tie_encoder_decoder": false,
188
+ "tie_word_embeddings": true,
189
+ "tokenizer_class": null,
190
+ "use_bidirectional_attention": false,
191
+ "use_cache": true,
192
+ "vocab_size": 262144
193
+ },
194
+ "vision_config": {
195
+ "_name_or_path": "",
196
+ "add_cross_attention": false,
197
+ "architectures": null,
198
+ "attention_dropout": 0.0,
199
+ "bos_token_id": null,
200
+ "chunk_size_feed_forward": 0,
201
+ "cross_attention_hidden_size": null,
202
+ "decoder_start_token_id": null,
203
+ "dropout_rate": 0.0,
204
+ "dtype": "bfloat16",
205
+ "eos_token_id": null,
206
+ "finetuning_task": null,
207
+ "hidden_act": "gelu_pytorch_tanh",
208
+ "hidden_size": 1152,
209
+ "id2label": {
210
+ "0": "LABEL_0",
211
+ "1": "LABEL_1"
212
+ },
213
+ "image_size": 896,
214
+ "intermediate_size": 4304,
215
+ "is_decoder": false,
216
+ "is_encoder_decoder": false,
217
+ "label2id": {
218
+ "LABEL_0": 0,
219
+ "LABEL_1": 1
220
+ },
221
+ "layer_norm_eps": 1e-06,
222
+ "model_type": "siglip_vision_model",
223
+ "num_attention_heads": 16,
224
+ "num_channels": 3,
225
+ "num_hidden_layers": 27,
226
+ "output_attentions": false,
227
+ "output_hidden_states": false,
228
+ "pad_token_id": null,
229
+ "patch_size": 14,
230
+ "prefix": null,
231
+ "problem_type": null,
232
+ "return_dict": true,
233
+ "sep_token_id": null,
234
+ "task_specific_params": null,
235
+ "tie_encoder_decoder": false,
236
+ "tie_word_embeddings": true,
237
+ "tokenizer_class": null,
238
+ "vision_use_head": false,
239
+ "vocab_size": 262144
240
+ },
241
+ "vocab_size": 262144
242
+ },
243
+ "eoi_token_index": 256000,
244
+ "eos_token_id": 1,
245
+ "image_token_index": 256001,
246
+ "initializer_range": 0.02,
247
+ "is_encoder_decoder": true,
248
+ "model_type": "t5gemma2",
249
+ "pad_token_id": 0,
250
+ "transformers_version": "5.0.0.dev0",
251
+ "vocab_size": 262144
252
+ }
T5-Gemma-2-4B/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": "5.0.0.dev0"
10
+ }
T5-Gemma-2-4B/model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70b70cc4289f699f601be45452fb76d90c1dbc810c8e3efcb7d3bb27d0fcd815
3
+ size 4960949936
T5-Gemma-2-4B/model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d22ece63fe00ddaadfa8f9cee178e95168b56a233b4c835183bef2d18fa061a
3
+ size 4991851208
T5-Gemma-2-4B/model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6af955d515b3bc84f74cdd40baf856bc28f985d3d762a59e3d9739a121492fe2
3
+ size 4960366008
T5-Gemma-2-4B/model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb13d4edbc893afb8fc7f710a7e0788ee920d1b456a50bc84e32637de663fefa
3
+ size 104868376
T5-Gemma-2-4B/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
T5-Gemma-2-4B/processor_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_processor": {
3
+ "do_convert_rgb": true,
4
+ "do_normalize": true,
5
+ "do_pan_and_scan": null,
6
+ "do_rescale": true,
7
+ "do_resize": true,
8
+ "image_mean": [
9
+ 0.5,
10
+ 0.5,
11
+ 0.5
12
+ ],
13
+ "image_processor_type": "Gemma3ImageProcessor",
14
+ "image_seq_length": 256,
15
+ "image_std": [
16
+ 0.5,
17
+ 0.5,
18
+ 0.5
19
+ ],
20
+ "pan_and_scan_max_num_crops": null,
21
+ "pan_and_scan_min_crop_size": null,
22
+ "pan_and_scan_min_ratio_to_activate": null,
23
+ "processor_class": "Gemma3Processor",
24
+ "resample": 2,
25
+ "rescale_factor": 0.00392156862745098,
26
+ "size": {
27
+ "height": 896,
28
+ "width": 896
29
+ }
30
+ },
31
+ "image_seq_length": 256,
32
+ "processor_class": "Gemma3Processor"
33
+ }
T5-Gemma-2-4B/special_tokens_map.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<eos>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "unk_token": {
27
+ "content": "<unk>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
T5-Gemma-2-4B/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c76318660cc87c3029535cc1596717df67132d231ce4ecf088ba2a8ec07c098
3
+ size 33384388
T5-Gemma-2-4B/tokenizer_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "additional_special_tokens": null,
5
+ "backend": "tokenizers",
6
+ "boi_token": "<start_of_image>",
7
+ "bos_token": "<bos>",
8
+ "clean_up_tokenization_spaces": false,
9
+ "eoi_token": "<end_of_image>",
10
+ "eos_token": "<eos>",
11
+ "image_token": "<image_soft_token>",
12
+ "is_local": true,
13
+ "mask_token": "<mask>",
14
+ "model_max_length": 1000000000000000019884624838656,
15
+ "model_specific_special_tokens": {
16
+ "boi_token": "<start_of_image>",
17
+ "eoi_token": "<end_of_image>",
18
+ "image_token": "<image_soft_token>"
19
+ },
20
+ "pad_token": "<pad>",
21
+ "padding_side": "right",
22
+ "processor_class": "Gemma3Processor",
23
+ "sp_model_kwargs": null,
24
+ "spaces_between_special_tokens": false,
25
+ "tokenizer_class": "GemmaTokenizer",
26
+ "unk_token": "<unk>",
27
+ "use_default_system_prompt": false
28
+ }