Hugging Face's logo

KaLM-Embedding
/
KaLM-Reranker-V1-Small

Text Ranking
Transformers
Safetensors
multilingual
t5gemma2
text2text-generation
reranker
encoder-decoder
FBNL
Retrieval
RAG
Model card Files
xet
 Community
Yuki131 commited on
Commit
de3d7ed
·
verified ·
1 Parent(s): 4cdc11a

Upload 4 files

Browse files
Files changed (5) hide show
  1. .gitattributes +1 -0
  2. config.json +239 -0
  3. generation_config.json +10 -0
  4. tokenizer.json +3 -0
  5. tokenizer_config.json +26 -0
.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 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5Gemma2ForConditionalGeneration"
4
+ ],
5
+ "bos_token_id": 2,
6
+ "classifier_dropout_rate": 0.0,
7
+ "decoder": {
8
+ "_sliding_window_pattern": 6,
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "bos_token_id": 2,
13
+ "dropout_rate": 0.0,
14
+ "dtype": "bfloat16",
15
+ "eos_token_id": 1,
16
+ "final_logit_softcapping": null,
17
+ "head_dim": 256,
18
+ "hidden_activation": "gelu_pytorch_tanh",
19
+ "hidden_size": 1152,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 6912,
22
+ "layer_types": [
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "full_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "full_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "full_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "full_attention",
47
+ "sliding_attention",
48
+ "sliding_attention"
49
+ ],
50
+ "max_position_embeddings": 32768,
51
+ "model_type": "t5gemma2_decoder",
52
+ "num_attention_heads": 4,
53
+ "num_hidden_layers": 26,
54
+ "num_key_value_heads": 1,
55
+ "pad_token_id": 0,
56
+ "query_pre_attn_scalar": 256,
57
+ "rms_norm_eps": 1e-06,
58
+ "rope_parameters": {
59
+ "full_attention": {
60
+ "factor": 8.0,
61
+ "rope_theta": 1000000,
62
+ "rope_type": "linear"
63
+ },
64
+ "sliding_attention": {
65
+ "rope_theta": 10000,
66
+ "rope_type": "default"
67
+ }
68
+ },
69
+ "sliding_window": 512,
70
+ "use_bidirectional_attention": false,
71
+ "use_cache": true,
72
+ "vocab_size": 262144
73
+ },
74
+ "dtype": "bfloat16",
75
+ "encoder": {
76
+ "attention_dropout": 0.0,
77
+ "boi_token_index": 255999,
78
+ "dropout_rate": 0.0,
79
+ "dtype": "bfloat16",
80
+ "eoi_token_index": 256000,
81
+ "image_token_index": 256001,
82
+ "initializer_range": 0.02,
83
+ "mm_tokens_per_image": 256,
84
+ "model_type": "t5gemma2_encoder",
85
+ "text_config": {
86
+ "_name_or_path": "",
87
+ "_sliding_window_pattern": 6,
88
+ "add_cross_attention": false,
89
+ "architectures": null,
90
+ "attention_bias": false,
91
+ "attention_dropout": 0.0,
92
+ "attn_logit_softcapping": null,
93
+ "bos_token_id": 2,
94
+ "chunk_size_feed_forward": 0,
95
+ "cross_attention_hidden_size": null,
96
+ "decoder_start_token_id": null,
97
+ "dropout_rate": 0.0,
98
+ "dtype": "bfloat16",
99
+ "eos_token_id": 1,
100
+ "final_logit_softcapping": null,
101
+ "finetuning_task": null,
102
+ "head_dim": 256,
103
+ "hidden_activation": "gelu_pytorch_tanh",
104
+ "hidden_size": 1152,
105
+ "id2label": {
106
+ "0": "LABEL_0",
107
+ "1": "LABEL_1"
108
+ },
109
+ "initializer_range": 0.02,
110
+ "intermediate_size": 6912,
111
+ "is_decoder": false,
112
+ "is_encoder_decoder": false,
113
+ "label2id": {
114
+ "LABEL_0": 0,
115
+ "LABEL_1": 1
116
+ },
117
+ "layer_types": [
118
+ "sliding_attention",
119
+ "sliding_attention",
120
+ "sliding_attention",
121
+ "sliding_attention",
122
+ "sliding_attention",
123
+ "full_attention",
124
+ "sliding_attention",
125
+ "sliding_attention",
126
+ "sliding_attention",
127
+ "sliding_attention",
128
+ "sliding_attention",
129
+ "full_attention",
130
+ "sliding_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
+ "sliding_attention",
141
+ "full_attention",
142
+ "sliding_attention",
143
+ "sliding_attention"
144
+ ],
145
+ "max_position_embeddings": 32768,
146
+ "model_type": "t5gemma2_text",
147
+ "num_attention_heads": 4,
148
+ "num_hidden_layers": 26,
149
+ "num_key_value_heads": 1,
150
+ "output_attentions": false,
151
+ "output_hidden_states": false,
152
+ "pad_token_id": 0,
153
+ "prefix": null,
154
+ "problem_type": null,
155
+ "query_pre_attn_scalar": 256,
156
+ "return_dict": true,
157
+ "rms_norm_eps": 1e-06,
158
+ "rope_parameters": {
159
+ "full_attention": {
160
+ "factor": 8.0,
161
+ "rope_theta": 1000000,
162
+ "rope_type": "linear"
163
+ },
164
+ "sliding_attention": {
165
+ "rope_theta": 10000,
166
+ "rope_type": "default"
167
+ }
168
+ },
169
+ "sep_token_id": null,
170
+ "sliding_window": 512,
171
+ "task_specific_params": null,
172
+ "tie_encoder_decoder": false,
173
+ "tie_word_embeddings": true,
174
+ "tokenizer_class": null,
175
+ "use_bidirectional_attention": false,
176
+ "use_cache": true,
177
+ "vocab_size": 262144
178
+ },
179
+ "tie_word_embeddings": true,
180
+ "vision_config": {
181
+ "_name_or_path": "",
182
+ "add_cross_attention": false,
183
+ "architectures": null,
184
+ "attention_dropout": 0.0,
185
+ "bos_token_id": null,
186
+ "chunk_size_feed_forward": 0,
187
+ "cross_attention_hidden_size": null,
188
+ "decoder_start_token_id": null,
189
+ "dropout_rate": 0.0,
190
+ "dtype": "bfloat16",
191
+ "eos_token_id": null,
192
+ "finetuning_task": null,
193
+ "hidden_act": "gelu_pytorch_tanh",
194
+ "hidden_size": 1152,
195
+ "id2label": {
196
+ "0": "LABEL_0",
197
+ "1": "LABEL_1"
198
+ },
199
+ "image_size": 896,
200
+ "intermediate_size": 4304,
201
+ "is_decoder": false,
202
+ "is_encoder_decoder": false,
203
+ "label2id": {
204
+ "LABEL_0": 0,
205
+ "LABEL_1": 1
206
+ },
207
+ "layer_norm_eps": 1e-06,
208
+ "model_type": "siglip_vision_model",
209
+ "num_attention_heads": 16,
210
+ "num_channels": 3,
211
+ "num_hidden_layers": 27,
212
+ "output_attentions": false,
213
+ "output_hidden_states": false,
214
+ "pad_token_id": null,
215
+ "patch_size": 14,
216
+ "prefix": null,
217
+ "problem_type": null,
218
+ "return_dict": true,
219
+ "sep_token_id": null,
220
+ "task_specific_params": null,
221
+ "tie_encoder_decoder": false,
222
+ "tie_word_embeddings": true,
223
+ "tokenizer_class": null,
224
+ "vision_use_head": false,
225
+ "vocab_size": 262144
226
+ },
227
+ "vocab_size": 262144
228
+ },
229
+ "eoi_token_index": 256000,
230
+ "eos_token_id": 1,
231
+ "image_token_index": 256001,
232
+ "initializer_range": 0.02,
233
+ "is_encoder_decoder": true,
234
+ "model_type": "t5gemma2",
235
+ "pad_token_id": 0,
236
+ "tie_word_embeddings": true,
237
+ "transformers_version": "5.3.0",
238
+ "vocab_size": 262144
239
+ }
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": "5.3.0"
10
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5b325224482ec441ec5fbe2a5ac08c3758e0f9605f6e54368e31f736fcfb01d
3
+ size 33384387
tokenizer_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "boi_token": "<start_of_image>",
4
+ "bos_token": "<bos>",
5
+ "clean_up_tokenization_spaces": false,
6
+ "eoi_token": "<end_of_image>",
7
+ "eos_token": "<eos>",
8
+ "extra_special_tokens": [],
9
+ "image_token": "<image_soft_token>",
10
+ "is_local": true,
11
+ "mask_token": "<mask>",
12
+ "model_max_length": 1000000000000000019884624838656,
13
+ "model_specific_special_tokens": {
14
+ "boi_token": "<start_of_image>",
15
+ "eoi_token": "<end_of_image>",
16
+ "image_token": "<image_soft_token>"
17
+ },
18
+ "pad_token": "<pad>",
19
+ "padding_side": "right",
20
+ "processor_class": "Gemma3Processor",
21
+ "sp_model_kwargs": null,
22
+ "spaces_between_special_tokens": false,
23
+ "tokenizer_class": "GemmaTokenizer",
24
+ "unk_token": "<unk>",
25
+ "use_default_system_prompt": false
26
+ }