Update config.json
Browse files- config.json +8 -49
config.json
CHANGED
|
@@ -3,19 +3,18 @@
|
|
| 3 |
"Gemma3ForConditionalGeneration"
|
| 4 |
],
|
| 5 |
"boi_token_index": 255999,
|
| 6 |
-
"bos_token_id": 2,
|
| 7 |
-
"dtype": "bfloat16",
|
| 8 |
"eoi_token_index": 256000,
|
| 9 |
-
"eos_token_id":
|
|
|
|
|
|
|
|
|
|
| 10 |
"image_token_index": 262144,
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
"mm_tokens_per_image": 256,
|
| 13 |
"model_type": "gemma3",
|
| 14 |
-
"pad_token_id": 0,
|
| 15 |
"quantization_config": {
|
| 16 |
"config_groups": {
|
| 17 |
"group_0": {
|
| 18 |
-
"format": "float-quantized",
|
| 19 |
"input_activations": {
|
| 20 |
"actorder": null,
|
| 21 |
"block_structure": null,
|
|
@@ -47,7 +46,7 @@
|
|
| 47 |
}
|
| 48 |
},
|
| 49 |
"format": "float-quantized",
|
| 50 |
-
"global_compression_ratio":
|
| 51 |
"ignore": [
|
| 52 |
"vision_tower.vision_model.encoder.layers.0.self_attn.k_proj",
|
| 53 |
"vision_tower.vision_model.encoder.layers.0.self_attn.v_proj",
|
|
@@ -215,57 +214,19 @@
|
|
| 215 |
],
|
| 216 |
"kv_cache_scheme": null,
|
| 217 |
"quant_method": "compressed-tensors",
|
| 218 |
-
"quantization_status": "compressed"
|
| 219 |
},
|
| 220 |
"text_config": {
|
| 221 |
-
"_sliding_window_pattern": 6,
|
| 222 |
"attention_bias": false,
|
| 223 |
"attention_dropout": 0.0,
|
| 224 |
"attn_logit_softcapping": null,
|
| 225 |
"cache_implementation": "hybrid",
|
| 226 |
-
"dtype": "bfloat16",
|
| 227 |
"final_logit_softcapping": null,
|
| 228 |
"head_dim": 256,
|
| 229 |
"hidden_activation": "gelu_pytorch_tanh",
|
| 230 |
"hidden_size": 2560,
|
| 231 |
"initializer_range": 0.02,
|
| 232 |
"intermediate_size": 10240,
|
| 233 |
-
"layer_types": [
|
| 234 |
-
"sliding_attention",
|
| 235 |
-
"sliding_attention",
|
| 236 |
-
"sliding_attention",
|
| 237 |
-
"sliding_attention",
|
| 238 |
-
"sliding_attention",
|
| 239 |
-
"full_attention",
|
| 240 |
-
"sliding_attention",
|
| 241 |
-
"sliding_attention",
|
| 242 |
-
"sliding_attention",
|
| 243 |
-
"sliding_attention",
|
| 244 |
-
"sliding_attention",
|
| 245 |
-
"full_attention",
|
| 246 |
-
"sliding_attention",
|
| 247 |
-
"sliding_attention",
|
| 248 |
-
"sliding_attention",
|
| 249 |
-
"sliding_attention",
|
| 250 |
-
"sliding_attention",
|
| 251 |
-
"full_attention",
|
| 252 |
-
"sliding_attention",
|
| 253 |
-
"sliding_attention",
|
| 254 |
-
"sliding_attention",
|
| 255 |
-
"sliding_attention",
|
| 256 |
-
"sliding_attention",
|
| 257 |
-
"full_attention",
|
| 258 |
-
"sliding_attention",
|
| 259 |
-
"sliding_attention",
|
| 260 |
-
"sliding_attention",
|
| 261 |
-
"sliding_attention",
|
| 262 |
-
"sliding_attention",
|
| 263 |
-
"full_attention",
|
| 264 |
-
"sliding_attention",
|
| 265 |
-
"sliding_attention",
|
| 266 |
-
"sliding_attention",
|
| 267 |
-
"sliding_attention"
|
| 268 |
-
],
|
| 269 |
"max_position_embeddings": 131072,
|
| 270 |
"model_type": "gemma3_text",
|
| 271 |
"num_attention_heads": 8,
|
|
@@ -281,15 +242,13 @@
|
|
| 281 |
"rope_theta": 1000000.0,
|
| 282 |
"sliding_window": 1024,
|
| 283 |
"sliding_window_pattern": 6,
|
| 284 |
-
"use_bidirectional_attention": false,
|
| 285 |
"use_cache": true,
|
| 286 |
"vocab_size": 262208
|
| 287 |
},
|
| 288 |
-
"
|
| 289 |
-
"
|
| 290 |
"vision_config": {
|
| 291 |
"attention_dropout": 0.0,
|
| 292 |
-
"dtype": "bfloat16",
|
| 293 |
"hidden_act": "gelu_pytorch_tanh",
|
| 294 |
"hidden_size": 1152,
|
| 295 |
"image_size": 896,
|
|
|
|
| 3 |
"Gemma3ForConditionalGeneration"
|
| 4 |
],
|
| 5 |
"boi_token_index": 255999,
|
|
|
|
|
|
|
| 6 |
"eoi_token_index": 256000,
|
| 7 |
+
"eos_token_id": [
|
| 8 |
+
1,
|
| 9 |
+
106
|
| 10 |
+
],
|
| 11 |
"image_token_index": 262144,
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
"mm_tokens_per_image": 256,
|
| 14 |
"model_type": "gemma3",
|
|
|
|
| 15 |
"quantization_config": {
|
| 16 |
"config_groups": {
|
| 17 |
"group_0": {
|
|
|
|
| 18 |
"input_activations": {
|
| 19 |
"actorder": null,
|
| 20 |
"block_structure": null,
|
|
|
|
| 46 |
}
|
| 47 |
},
|
| 48 |
"format": "float-quantized",
|
| 49 |
+
"global_compression_ratio": 1.1950546068196743,
|
| 50 |
"ignore": [
|
| 51 |
"vision_tower.vision_model.encoder.layers.0.self_attn.k_proj",
|
| 52 |
"vision_tower.vision_model.encoder.layers.0.self_attn.v_proj",
|
|
|
|
| 214 |
],
|
| 215 |
"kv_cache_scheme": null,
|
| 216 |
"quant_method": "compressed-tensors",
|
| 217 |
+
"quantization_status": "compressed"
|
| 218 |
},
|
| 219 |
"text_config": {
|
|
|
|
| 220 |
"attention_bias": false,
|
| 221 |
"attention_dropout": 0.0,
|
| 222 |
"attn_logit_softcapping": null,
|
| 223 |
"cache_implementation": "hybrid",
|
|
|
|
| 224 |
"final_logit_softcapping": null,
|
| 225 |
"head_dim": 256,
|
| 226 |
"hidden_activation": "gelu_pytorch_tanh",
|
| 227 |
"hidden_size": 2560,
|
| 228 |
"initializer_range": 0.02,
|
| 229 |
"intermediate_size": 10240,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 230 |
"max_position_embeddings": 131072,
|
| 231 |
"model_type": "gemma3_text",
|
| 232 |
"num_attention_heads": 8,
|
|
|
|
| 242 |
"rope_theta": 1000000.0,
|
| 243 |
"sliding_window": 1024,
|
| 244 |
"sliding_window_pattern": 6,
|
|
|
|
| 245 |
"use_cache": true,
|
| 246 |
"vocab_size": 262208
|
| 247 |
},
|
| 248 |
+
"torch_dtype": "bfloat16",
|
| 249 |
+
"transformers_version": "4.50.0",
|
| 250 |
"vision_config": {
|
| 251 |
"attention_dropout": 0.0,
|
|
|
|
| 252 |
"hidden_act": "gelu_pytorch_tanh",
|
| 253 |
"hidden_size": 1152,
|
| 254 |
"image_size": 896,
|