Upload folder using huggingface_hub
Browse files- _config.json +197 -0
- chat_template.jinja +0 -9
- config.json +377 -6
- model.safetensors +2 -2
- processor_config.json +1 -1
_config.json
ADDED
|
@@ -0,0 +1,197 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma4ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"audio_config": {
|
| 6 |
+
"_name_or_path": "",
|
| 7 |
+
"architectures": null,
|
| 8 |
+
"attention_chunk_size": 12,
|
| 9 |
+
"attention_context_left": 13,
|
| 10 |
+
"attention_context_right": 0,
|
| 11 |
+
"attention_invalid_logits_value": -1000000000.0,
|
| 12 |
+
"attention_logit_cap": 50.0,
|
| 13 |
+
"chunk_size_feed_forward": 0,
|
| 14 |
+
"conv_kernel_size": 5,
|
| 15 |
+
"dtype": "bfloat16",
|
| 16 |
+
"gradient_clipping": 10000000000.0,
|
| 17 |
+
"hidden_act": "silu",
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
+
"id2label": {
|
| 20 |
+
"0": "LABEL_0",
|
| 21 |
+
"1": "LABEL_1"
|
| 22 |
+
},
|
| 23 |
+
"initializer_range": 0.02,
|
| 24 |
+
"is_encoder_decoder": false,
|
| 25 |
+
"label2id": {
|
| 26 |
+
"LABEL_0": 0,
|
| 27 |
+
"LABEL_1": 1
|
| 28 |
+
},
|
| 29 |
+
"model_type": "gemma4_audio",
|
| 30 |
+
"num_attention_heads": 8,
|
| 31 |
+
"num_hidden_layers": 12,
|
| 32 |
+
"output_attentions": false,
|
| 33 |
+
"output_hidden_states": false,
|
| 34 |
+
"output_proj_dims": 1536,
|
| 35 |
+
"problem_type": null,
|
| 36 |
+
"residual_weight": 0.5,
|
| 37 |
+
"return_dict": true,
|
| 38 |
+
"rms_norm_eps": 1e-06,
|
| 39 |
+
"subsampling_conv_channels": [
|
| 40 |
+
128,
|
| 41 |
+
32
|
| 42 |
+
],
|
| 43 |
+
"use_clipped_linears": true
|
| 44 |
+
},
|
| 45 |
+
"audio_token_id": 258881,
|
| 46 |
+
"boa_token_id": 256000,
|
| 47 |
+
"boi_token_id": 255999,
|
| 48 |
+
"dtype": "bfloat16",
|
| 49 |
+
"eoa_token_id": 258883,
|
| 50 |
+
"eoa_token_index": 258883,
|
| 51 |
+
"eoi_token_id": 258882,
|
| 52 |
+
"eos_token_id": [
|
| 53 |
+
1,
|
| 54 |
+
106
|
| 55 |
+
],
|
| 56 |
+
"image_token_id": 258880,
|
| 57 |
+
"initializer_range": 0.02,
|
| 58 |
+
"model_type": "gemma4",
|
| 59 |
+
"text_config": {
|
| 60 |
+
"attention_bias": false,
|
| 61 |
+
"attention_dropout": 0.0,
|
| 62 |
+
"attention_k_eq_v": false,
|
| 63 |
+
"bos_token_id": 2,
|
| 64 |
+
"dtype": "bfloat16",
|
| 65 |
+
"enable_moe_block": false,
|
| 66 |
+
"eos_token_id": 1,
|
| 67 |
+
"expert_intermediate_size": null,
|
| 68 |
+
"final_logit_softcapping": 30.0,
|
| 69 |
+
"global_head_dim": 512,
|
| 70 |
+
"head_dim": 256,
|
| 71 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 72 |
+
"hidden_size": 2560,
|
| 73 |
+
"hidden_size_per_layer_input": 256,
|
| 74 |
+
"initializer_range": 0.02,
|
| 75 |
+
"intermediate_size": 10240,
|
| 76 |
+
"layer_types": [
|
| 77 |
+
"sliding_attention",
|
| 78 |
+
"sliding_attention",
|
| 79 |
+
"sliding_attention",
|
| 80 |
+
"sliding_attention",
|
| 81 |
+
"sliding_attention",
|
| 82 |
+
"full_attention",
|
| 83 |
+
"sliding_attention",
|
| 84 |
+
"sliding_attention",
|
| 85 |
+
"sliding_attention",
|
| 86 |
+
"sliding_attention",
|
| 87 |
+
"sliding_attention",
|
| 88 |
+
"full_attention",
|
| 89 |
+
"sliding_attention",
|
| 90 |
+
"sliding_attention",
|
| 91 |
+
"sliding_attention",
|
| 92 |
+
"sliding_attention",
|
| 93 |
+
"sliding_attention",
|
| 94 |
+
"full_attention",
|
| 95 |
+
"sliding_attention",
|
| 96 |
+
"sliding_attention",
|
| 97 |
+
"sliding_attention",
|
| 98 |
+
"sliding_attention",
|
| 99 |
+
"sliding_attention",
|
| 100 |
+
"full_attention",
|
| 101 |
+
"sliding_attention",
|
| 102 |
+
"sliding_attention",
|
| 103 |
+
"sliding_attention",
|
| 104 |
+
"sliding_attention",
|
| 105 |
+
"sliding_attention",
|
| 106 |
+
"full_attention",
|
| 107 |
+
"sliding_attention",
|
| 108 |
+
"sliding_attention",
|
| 109 |
+
"sliding_attention",
|
| 110 |
+
"sliding_attention",
|
| 111 |
+
"sliding_attention",
|
| 112 |
+
"full_attention",
|
| 113 |
+
"sliding_attention",
|
| 114 |
+
"sliding_attention",
|
| 115 |
+
"sliding_attention",
|
| 116 |
+
"sliding_attention",
|
| 117 |
+
"sliding_attention",
|
| 118 |
+
"full_attention"
|
| 119 |
+
],
|
| 120 |
+
"max_position_embeddings": 131072,
|
| 121 |
+
"model_type": "gemma4_text",
|
| 122 |
+
"num_attention_heads": 8,
|
| 123 |
+
"num_experts": null,
|
| 124 |
+
"num_global_key_value_heads": null,
|
| 125 |
+
"num_hidden_layers": 42,
|
| 126 |
+
"num_key_value_heads": 2,
|
| 127 |
+
"num_kv_shared_layers": 18,
|
| 128 |
+
"pad_token_id": 0,
|
| 129 |
+
"rms_norm_eps": 1e-06,
|
| 130 |
+
"rope_parameters": {
|
| 131 |
+
"full_attention": {
|
| 132 |
+
"partial_rotary_factor": 0.25,
|
| 133 |
+
"rope_theta": 1000000.0,
|
| 134 |
+
"rope_type": "proportional"
|
| 135 |
+
},
|
| 136 |
+
"sliding_attention": {
|
| 137 |
+
"rope_theta": 10000.0,
|
| 138 |
+
"rope_type": "default"
|
| 139 |
+
}
|
| 140 |
+
},
|
| 141 |
+
"sliding_window": 512,
|
| 142 |
+
"tie_word_embeddings": true,
|
| 143 |
+
"top_k_experts": null,
|
| 144 |
+
"use_bidirectional_attention": null,
|
| 145 |
+
"use_cache": true,
|
| 146 |
+
"use_double_wide_mlp": false,
|
| 147 |
+
"vocab_size": 262144,
|
| 148 |
+
"vocab_size_per_layer_input": 262144
|
| 149 |
+
},
|
| 150 |
+
"tie_word_embeddings": true,
|
| 151 |
+
"transformers_version": "5.5.0.dev0",
|
| 152 |
+
"video_token_id": 258884,
|
| 153 |
+
"vision_config": {
|
| 154 |
+
"_name_or_path": "",
|
| 155 |
+
"architectures": null,
|
| 156 |
+
"attention_bias": false,
|
| 157 |
+
"attention_dropout": 0.0,
|
| 158 |
+
"chunk_size_feed_forward": 0,
|
| 159 |
+
"default_output_length": 280,
|
| 160 |
+
"dtype": "bfloat16",
|
| 161 |
+
"global_head_dim": 64,
|
| 162 |
+
"head_dim": 64,
|
| 163 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 164 |
+
"hidden_size": 768,
|
| 165 |
+
"id2label": {
|
| 166 |
+
"0": "LABEL_0",
|
| 167 |
+
"1": "LABEL_1"
|
| 168 |
+
},
|
| 169 |
+
"initializer_range": 0.02,
|
| 170 |
+
"intermediate_size": 3072,
|
| 171 |
+
"is_encoder_decoder": false,
|
| 172 |
+
"label2id": {
|
| 173 |
+
"LABEL_0": 0,
|
| 174 |
+
"LABEL_1": 1
|
| 175 |
+
},
|
| 176 |
+
"max_position_embeddings": 131072,
|
| 177 |
+
"model_type": "gemma4_vision",
|
| 178 |
+
"num_attention_heads": 12,
|
| 179 |
+
"num_hidden_layers": 16,
|
| 180 |
+
"num_key_value_heads": 12,
|
| 181 |
+
"output_attentions": false,
|
| 182 |
+
"output_hidden_states": false,
|
| 183 |
+
"patch_size": 16,
|
| 184 |
+
"pooling_kernel_size": 3,
|
| 185 |
+
"position_embedding_size": 10240,
|
| 186 |
+
"problem_type": null,
|
| 187 |
+
"return_dict": true,
|
| 188 |
+
"rms_norm_eps": 1e-06,
|
| 189 |
+
"rope_parameters": {
|
| 190 |
+
"rope_theta": 100.0,
|
| 191 |
+
"rope_type": "default"
|
| 192 |
+
},
|
| 193 |
+
"standardize": false,
|
| 194 |
+
"use_clipped_linears": true
|
| 195 |
+
},
|
| 196 |
+
"vision_soft_tokens_per_image": 280
|
| 197 |
+
}
|
chat_template.jinja
CHANGED
|
@@ -295,15 +295,6 @@
|
|
| 295 |
{%- endif -%}
|
| 296 |
{%- endfor -%}
|
| 297 |
{{- format_tool_response_block(ns_tname.name, ns_txt.s) -}}
|
| 298 |
-
{%- for part in tool_body -%}
|
| 299 |
-
{%- if part.get('type') == 'image' -%}
|
| 300 |
-
{{- '<|image|>' -}}
|
| 301 |
-
{%- elif part.get('type') == 'audio' -%}
|
| 302 |
-
{{- '<|audio|>' -}}
|
| 303 |
-
{%- elif part.get('type') == 'video' -%}
|
| 304 |
-
{{- '<|video|>' -}}
|
| 305 |
-
{%- endif -%}
|
| 306 |
-
{%- endfor -%}
|
| 307 |
{%- else -%}
|
| 308 |
{{- format_tool_response_block(ns_tname.name, tool_body) -}}
|
| 309 |
{%- endif -%}
|
|
|
|
| 295 |
{%- endif -%}
|
| 296 |
{%- endfor -%}
|
| 297 |
{{- format_tool_response_block(ns_tname.name, ns_txt.s) -}}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 298 |
{%- else -%}
|
| 299 |
{{- format_tool_response_block(ns_tname.name, tool_body) -}}
|
| 300 |
{%- endif -%}
|
config.json
CHANGED
|
@@ -12,7 +12,7 @@
|
|
| 12 |
"attention_logit_cap": 50.0,
|
| 13 |
"chunk_size_feed_forward": 0,
|
| 14 |
"conv_kernel_size": 5,
|
| 15 |
-
"dtype": "
|
| 16 |
"gradient_clipping": 10000000000.0,
|
| 17 |
"hidden_act": "silu",
|
| 18 |
"hidden_size": 1024,
|
|
@@ -45,7 +45,7 @@
|
|
| 45 |
"audio_token_id": 258881,
|
| 46 |
"boa_token_id": 256000,
|
| 47 |
"boi_token_id": 255999,
|
| 48 |
-
"dtype": "
|
| 49 |
"eoa_token_id": 258883,
|
| 50 |
"eoa_token_index": 258883,
|
| 51 |
"eoi_token_id": 258882,
|
|
@@ -56,12 +56,382 @@
|
|
| 56 |
"image_token_id": 258880,
|
| 57 |
"initializer_range": 0.02,
|
| 58 |
"model_type": "gemma4",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
"text_config": {
|
| 60 |
"attention_bias": false,
|
| 61 |
"attention_dropout": 0.0,
|
| 62 |
"attention_k_eq_v": false,
|
| 63 |
"bos_token_id": 2,
|
| 64 |
-
"dtype": "
|
| 65 |
"enable_moe_block": false,
|
| 66 |
"eos_token_id": 1,
|
| 67 |
"expert_intermediate_size": null,
|
|
@@ -119,6 +489,7 @@
|
|
| 119 |
],
|
| 120 |
"max_position_embeddings": 131072,
|
| 121 |
"model_type": "gemma4_text",
|
|
|
|
| 122 |
"num_attention_heads": 8,
|
| 123 |
"num_experts": null,
|
| 124 |
"num_global_key_value_heads": null,
|
|
@@ -148,7 +519,7 @@
|
|
| 148 |
"vocab_size_per_layer_input": 262144
|
| 149 |
},
|
| 150 |
"tie_word_embeddings": true,
|
| 151 |
-
"transformers_version": "5.
|
| 152 |
"video_token_id": 258884,
|
| 153 |
"vision_config": {
|
| 154 |
"_name_or_path": "",
|
|
@@ -157,7 +528,7 @@
|
|
| 157 |
"attention_dropout": 0.0,
|
| 158 |
"chunk_size_feed_forward": 0,
|
| 159 |
"default_output_length": 280,
|
| 160 |
-
"dtype": "
|
| 161 |
"global_head_dim": 64,
|
| 162 |
"head_dim": 64,
|
| 163 |
"hidden_activation": "gelu_pytorch_tanh",
|
|
@@ -194,4 +565,4 @@
|
|
| 194 |
"use_clipped_linears": true
|
| 195 |
},
|
| 196 |
"vision_soft_tokens_per_image": 280
|
| 197 |
-
}
|
|
|
|
| 12 |
"attention_logit_cap": 50.0,
|
| 13 |
"chunk_size_feed_forward": 0,
|
| 14 |
"conv_kernel_size": 5,
|
| 15 |
+
"dtype": "float16",
|
| 16 |
"gradient_clipping": 10000000000.0,
|
| 17 |
"hidden_act": "silu",
|
| 18 |
"hidden_size": 1024,
|
|
|
|
| 45 |
"audio_token_id": 258881,
|
| 46 |
"boa_token_id": 256000,
|
| 47 |
"boi_token_id": 255999,
|
| 48 |
+
"dtype": "float16",
|
| 49 |
"eoa_token_id": 258883,
|
| 50 |
"eoa_token_index": 258883,
|
| 51 |
"eoi_token_id": 258882,
|
|
|
|
| 56 |
"image_token_id": 258880,
|
| 57 |
"initializer_range": 0.02,
|
| 58 |
"model_type": "gemma4",
|
| 59 |
+
"quantization_config": {
|
| 60 |
+
"config_groups": {
|
| 61 |
+
"group_0": {
|
| 62 |
+
"format": "pack-quantized",
|
| 63 |
+
"input_activations": null,
|
| 64 |
+
"output_activations": null,
|
| 65 |
+
"targets": [
|
| 66 |
+
"Linear"
|
| 67 |
+
],
|
| 68 |
+
"weights": {
|
| 69 |
+
"actorder": null,
|
| 70 |
+
"block_structure": null,
|
| 71 |
+
"dynamic": false,
|
| 72 |
+
"group_size": 32,
|
| 73 |
+
"num_bits": 4,
|
| 74 |
+
"observer": "mse",
|
| 75 |
+
"observer_kwargs": {},
|
| 76 |
+
"scale_dtype": null,
|
| 77 |
+
"strategy": "group",
|
| 78 |
+
"symmetric": false,
|
| 79 |
+
"type": "int",
|
| 80 |
+
"zp_dtype": "torch.int8"
|
| 81 |
+
}
|
| 82 |
+
}
|
| 83 |
+
},
|
| 84 |
+
"format": "pack-quantized",
|
| 85 |
+
"global_compression_ratio": null,
|
| 86 |
+
"ignore": [
|
| 87 |
+
"model.vision_tower.patch_embedder.input_proj",
|
| 88 |
+
"model.vision_tower.encoder.layers.0.self_attn.q_proj.linear",
|
| 89 |
+
"model.vision_tower.encoder.layers.0.self_attn.k_proj.linear",
|
| 90 |
+
"model.vision_tower.encoder.layers.0.self_attn.v_proj.linear",
|
| 91 |
+
"model.vision_tower.encoder.layers.0.self_attn.o_proj.linear",
|
| 92 |
+
"model.vision_tower.encoder.layers.0.mlp.gate_proj.linear",
|
| 93 |
+
"model.vision_tower.encoder.layers.0.mlp.up_proj.linear",
|
| 94 |
+
"model.vision_tower.encoder.layers.0.mlp.down_proj.linear",
|
| 95 |
+
"model.vision_tower.encoder.layers.1.self_attn.q_proj.linear",
|
| 96 |
+
"model.vision_tower.encoder.layers.1.self_attn.k_proj.linear",
|
| 97 |
+
"model.vision_tower.encoder.layers.1.self_attn.v_proj.linear",
|
| 98 |
+
"model.vision_tower.encoder.layers.1.self_attn.o_proj.linear",
|
| 99 |
+
"model.vision_tower.encoder.layers.1.mlp.gate_proj.linear",
|
| 100 |
+
"model.vision_tower.encoder.layers.1.mlp.up_proj.linear",
|
| 101 |
+
"model.vision_tower.encoder.layers.1.mlp.down_proj.linear",
|
| 102 |
+
"model.vision_tower.encoder.layers.2.self_attn.q_proj.linear",
|
| 103 |
+
"model.vision_tower.encoder.layers.2.self_attn.k_proj.linear",
|
| 104 |
+
"model.vision_tower.encoder.layers.2.self_attn.v_proj.linear",
|
| 105 |
+
"model.vision_tower.encoder.layers.2.self_attn.o_proj.linear",
|
| 106 |
+
"model.vision_tower.encoder.layers.2.mlp.gate_proj.linear",
|
| 107 |
+
"model.vision_tower.encoder.layers.2.mlp.up_proj.linear",
|
| 108 |
+
"model.vision_tower.encoder.layers.2.mlp.down_proj.linear",
|
| 109 |
+
"model.vision_tower.encoder.layers.3.self_attn.q_proj.linear",
|
| 110 |
+
"model.vision_tower.encoder.layers.3.self_attn.k_proj.linear",
|
| 111 |
+
"model.vision_tower.encoder.layers.3.self_attn.v_proj.linear",
|
| 112 |
+
"model.vision_tower.encoder.layers.3.self_attn.o_proj.linear",
|
| 113 |
+
"model.vision_tower.encoder.layers.3.mlp.gate_proj.linear",
|
| 114 |
+
"model.vision_tower.encoder.layers.3.mlp.up_proj.linear",
|
| 115 |
+
"model.vision_tower.encoder.layers.3.mlp.down_proj.linear",
|
| 116 |
+
"model.vision_tower.encoder.layers.4.self_attn.q_proj.linear",
|
| 117 |
+
"model.vision_tower.encoder.layers.4.self_attn.k_proj.linear",
|
| 118 |
+
"model.vision_tower.encoder.layers.4.self_attn.v_proj.linear",
|
| 119 |
+
"model.vision_tower.encoder.layers.4.self_attn.o_proj.linear",
|
| 120 |
+
"model.vision_tower.encoder.layers.4.mlp.gate_proj.linear",
|
| 121 |
+
"model.vision_tower.encoder.layers.4.mlp.up_proj.linear",
|
| 122 |
+
"model.vision_tower.encoder.layers.4.mlp.down_proj.linear",
|
| 123 |
+
"model.vision_tower.encoder.layers.5.self_attn.q_proj.linear",
|
| 124 |
+
"model.vision_tower.encoder.layers.5.self_attn.k_proj.linear",
|
| 125 |
+
"model.vision_tower.encoder.layers.5.self_attn.v_proj.linear",
|
| 126 |
+
"model.vision_tower.encoder.layers.5.self_attn.o_proj.linear",
|
| 127 |
+
"model.vision_tower.encoder.layers.5.mlp.gate_proj.linear",
|
| 128 |
+
"model.vision_tower.encoder.layers.5.mlp.up_proj.linear",
|
| 129 |
+
"model.vision_tower.encoder.layers.5.mlp.down_proj.linear",
|
| 130 |
+
"model.vision_tower.encoder.layers.6.self_attn.q_proj.linear",
|
| 131 |
+
"model.vision_tower.encoder.layers.6.self_attn.k_proj.linear",
|
| 132 |
+
"model.vision_tower.encoder.layers.6.self_attn.v_proj.linear",
|
| 133 |
+
"model.vision_tower.encoder.layers.6.self_attn.o_proj.linear",
|
| 134 |
+
"model.vision_tower.encoder.layers.6.mlp.gate_proj.linear",
|
| 135 |
+
"model.vision_tower.encoder.layers.6.mlp.up_proj.linear",
|
| 136 |
+
"model.vision_tower.encoder.layers.6.mlp.down_proj.linear",
|
| 137 |
+
"model.vision_tower.encoder.layers.7.self_attn.q_proj.linear",
|
| 138 |
+
"model.vision_tower.encoder.layers.7.self_attn.k_proj.linear",
|
| 139 |
+
"model.vision_tower.encoder.layers.7.self_attn.v_proj.linear",
|
| 140 |
+
"model.vision_tower.encoder.layers.7.self_attn.o_proj.linear",
|
| 141 |
+
"model.vision_tower.encoder.layers.7.mlp.gate_proj.linear",
|
| 142 |
+
"model.vision_tower.encoder.layers.7.mlp.up_proj.linear",
|
| 143 |
+
"model.vision_tower.encoder.layers.7.mlp.down_proj.linear",
|
| 144 |
+
"model.vision_tower.encoder.layers.8.self_attn.q_proj.linear",
|
| 145 |
+
"model.vision_tower.encoder.layers.8.self_attn.k_proj.linear",
|
| 146 |
+
"model.vision_tower.encoder.layers.8.self_attn.v_proj.linear",
|
| 147 |
+
"model.vision_tower.encoder.layers.8.self_attn.o_proj.linear",
|
| 148 |
+
"model.vision_tower.encoder.layers.8.mlp.gate_proj.linear",
|
| 149 |
+
"model.vision_tower.encoder.layers.8.mlp.up_proj.linear",
|
| 150 |
+
"model.vision_tower.encoder.layers.8.mlp.down_proj.linear",
|
| 151 |
+
"model.vision_tower.encoder.layers.9.self_attn.q_proj.linear",
|
| 152 |
+
"model.vision_tower.encoder.layers.9.self_attn.k_proj.linear",
|
| 153 |
+
"model.vision_tower.encoder.layers.9.self_attn.v_proj.linear",
|
| 154 |
+
"model.vision_tower.encoder.layers.9.self_attn.o_proj.linear",
|
| 155 |
+
"model.vision_tower.encoder.layers.9.mlp.gate_proj.linear",
|
| 156 |
+
"model.vision_tower.encoder.layers.9.mlp.up_proj.linear",
|
| 157 |
+
"model.vision_tower.encoder.layers.9.mlp.down_proj.linear",
|
| 158 |
+
"model.vision_tower.encoder.layers.10.self_attn.q_proj.linear",
|
| 159 |
+
"model.vision_tower.encoder.layers.10.self_attn.k_proj.linear",
|
| 160 |
+
"model.vision_tower.encoder.layers.10.self_attn.v_proj.linear",
|
| 161 |
+
"model.vision_tower.encoder.layers.10.self_attn.o_proj.linear",
|
| 162 |
+
"model.vision_tower.encoder.layers.10.mlp.gate_proj.linear",
|
| 163 |
+
"model.vision_tower.encoder.layers.10.mlp.up_proj.linear",
|
| 164 |
+
"model.vision_tower.encoder.layers.10.mlp.down_proj.linear",
|
| 165 |
+
"model.vision_tower.encoder.layers.11.self_attn.q_proj.linear",
|
| 166 |
+
"model.vision_tower.encoder.layers.11.self_attn.k_proj.linear",
|
| 167 |
+
"model.vision_tower.encoder.layers.11.self_attn.v_proj.linear",
|
| 168 |
+
"model.vision_tower.encoder.layers.11.self_attn.o_proj.linear",
|
| 169 |
+
"model.vision_tower.encoder.layers.11.mlp.gate_proj.linear",
|
| 170 |
+
"model.vision_tower.encoder.layers.11.mlp.up_proj.linear",
|
| 171 |
+
"model.vision_tower.encoder.layers.11.mlp.down_proj.linear",
|
| 172 |
+
"model.vision_tower.encoder.layers.12.self_attn.q_proj.linear",
|
| 173 |
+
"model.vision_tower.encoder.layers.12.self_attn.k_proj.linear",
|
| 174 |
+
"model.vision_tower.encoder.layers.12.self_attn.v_proj.linear",
|
| 175 |
+
"model.vision_tower.encoder.layers.12.self_attn.o_proj.linear",
|
| 176 |
+
"model.vision_tower.encoder.layers.12.mlp.gate_proj.linear",
|
| 177 |
+
"model.vision_tower.encoder.layers.12.mlp.up_proj.linear",
|
| 178 |
+
"model.vision_tower.encoder.layers.12.mlp.down_proj.linear",
|
| 179 |
+
"model.vision_tower.encoder.layers.13.self_attn.q_proj.linear",
|
| 180 |
+
"model.vision_tower.encoder.layers.13.self_attn.k_proj.linear",
|
| 181 |
+
"model.vision_tower.encoder.layers.13.self_attn.v_proj.linear",
|
| 182 |
+
"model.vision_tower.encoder.layers.13.self_attn.o_proj.linear",
|
| 183 |
+
"model.vision_tower.encoder.layers.13.mlp.gate_proj.linear",
|
| 184 |
+
"model.vision_tower.encoder.layers.13.mlp.up_proj.linear",
|
| 185 |
+
"model.vision_tower.encoder.layers.13.mlp.down_proj.linear",
|
| 186 |
+
"model.vision_tower.encoder.layers.14.self_attn.q_proj.linear",
|
| 187 |
+
"model.vision_tower.encoder.layers.14.self_attn.k_proj.linear",
|
| 188 |
+
"model.vision_tower.encoder.layers.14.self_attn.v_proj.linear",
|
| 189 |
+
"model.vision_tower.encoder.layers.14.self_attn.o_proj.linear",
|
| 190 |
+
"model.vision_tower.encoder.layers.14.mlp.gate_proj.linear",
|
| 191 |
+
"model.vision_tower.encoder.layers.14.mlp.up_proj.linear",
|
| 192 |
+
"model.vision_tower.encoder.layers.14.mlp.down_proj.linear",
|
| 193 |
+
"model.vision_tower.encoder.layers.15.self_attn.q_proj.linear",
|
| 194 |
+
"model.vision_tower.encoder.layers.15.self_attn.k_proj.linear",
|
| 195 |
+
"model.vision_tower.encoder.layers.15.self_attn.v_proj.linear",
|
| 196 |
+
"model.vision_tower.encoder.layers.15.self_attn.o_proj.linear",
|
| 197 |
+
"model.vision_tower.encoder.layers.15.mlp.gate_proj.linear",
|
| 198 |
+
"model.vision_tower.encoder.layers.15.mlp.up_proj.linear",
|
| 199 |
+
"model.vision_tower.encoder.layers.15.mlp.down_proj.linear",
|
| 200 |
+
"model.language_model.layers.0.per_layer_input_gate",
|
| 201 |
+
"model.language_model.layers.0.per_layer_projection",
|
| 202 |
+
"model.language_model.layers.1.per_layer_input_gate",
|
| 203 |
+
"model.language_model.layers.1.per_layer_projection",
|
| 204 |
+
"model.language_model.layers.2.per_layer_input_gate",
|
| 205 |
+
"model.language_model.layers.2.per_layer_projection",
|
| 206 |
+
"model.language_model.layers.3.per_layer_input_gate",
|
| 207 |
+
"model.language_model.layers.3.per_layer_projection",
|
| 208 |
+
"model.language_model.layers.4.per_layer_input_gate",
|
| 209 |
+
"model.language_model.layers.4.per_layer_projection",
|
| 210 |
+
"model.language_model.layers.5.per_layer_input_gate",
|
| 211 |
+
"model.language_model.layers.5.per_layer_projection",
|
| 212 |
+
"model.language_model.layers.6.per_layer_input_gate",
|
| 213 |
+
"model.language_model.layers.6.per_layer_projection",
|
| 214 |
+
"model.language_model.layers.7.per_layer_input_gate",
|
| 215 |
+
"model.language_model.layers.7.per_layer_projection",
|
| 216 |
+
"model.language_model.layers.8.per_layer_input_gate",
|
| 217 |
+
"model.language_model.layers.8.per_layer_projection",
|
| 218 |
+
"model.language_model.layers.9.per_layer_input_gate",
|
| 219 |
+
"model.language_model.layers.9.per_layer_projection",
|
| 220 |
+
"model.language_model.layers.10.per_layer_input_gate",
|
| 221 |
+
"model.language_model.layers.10.per_layer_projection",
|
| 222 |
+
"model.language_model.layers.11.per_layer_input_gate",
|
| 223 |
+
"model.language_model.layers.11.per_layer_projection",
|
| 224 |
+
"model.language_model.layers.12.per_layer_input_gate",
|
| 225 |
+
"model.language_model.layers.12.per_layer_projection",
|
| 226 |
+
"model.language_model.layers.13.per_layer_input_gate",
|
| 227 |
+
"model.language_model.layers.13.per_layer_projection",
|
| 228 |
+
"model.language_model.layers.14.per_layer_input_gate",
|
| 229 |
+
"model.language_model.layers.14.per_layer_projection",
|
| 230 |
+
"model.language_model.layers.15.per_layer_input_gate",
|
| 231 |
+
"model.language_model.layers.15.per_layer_projection",
|
| 232 |
+
"model.language_model.layers.16.per_layer_input_gate",
|
| 233 |
+
"model.language_model.layers.16.per_layer_projection",
|
| 234 |
+
"model.language_model.layers.17.per_layer_input_gate",
|
| 235 |
+
"model.language_model.layers.17.per_layer_projection",
|
| 236 |
+
"model.language_model.layers.18.per_layer_input_gate",
|
| 237 |
+
"model.language_model.layers.18.per_layer_projection",
|
| 238 |
+
"model.language_model.layers.19.per_layer_input_gate",
|
| 239 |
+
"model.language_model.layers.19.per_layer_projection",
|
| 240 |
+
"model.language_model.layers.20.per_layer_input_gate",
|
| 241 |
+
"model.language_model.layers.20.per_layer_projection",
|
| 242 |
+
"model.language_model.layers.21.per_layer_input_gate",
|
| 243 |
+
"model.language_model.layers.21.per_layer_projection",
|
| 244 |
+
"model.language_model.layers.22.per_layer_input_gate",
|
| 245 |
+
"model.language_model.layers.22.per_layer_projection",
|
| 246 |
+
"model.language_model.layers.23.per_layer_input_gate",
|
| 247 |
+
"model.language_model.layers.23.per_layer_projection",
|
| 248 |
+
"model.language_model.layers.24.per_layer_input_gate",
|
| 249 |
+
"model.language_model.layers.24.per_layer_projection",
|
| 250 |
+
"model.language_model.layers.25.per_layer_input_gate",
|
| 251 |
+
"model.language_model.layers.25.per_layer_projection",
|
| 252 |
+
"model.language_model.layers.26.per_layer_input_gate",
|
| 253 |
+
"model.language_model.layers.26.per_layer_projection",
|
| 254 |
+
"model.language_model.layers.27.per_layer_input_gate",
|
| 255 |
+
"model.language_model.layers.27.per_layer_projection",
|
| 256 |
+
"model.language_model.layers.28.per_layer_input_gate",
|
| 257 |
+
"model.language_model.layers.28.per_layer_projection",
|
| 258 |
+
"model.language_model.layers.29.per_layer_input_gate",
|
| 259 |
+
"model.language_model.layers.29.per_layer_projection",
|
| 260 |
+
"model.language_model.layers.30.per_layer_input_gate",
|
| 261 |
+
"model.language_model.layers.30.per_layer_projection",
|
| 262 |
+
"model.language_model.layers.31.per_layer_input_gate",
|
| 263 |
+
"model.language_model.layers.31.per_layer_projection",
|
| 264 |
+
"model.language_model.layers.32.per_layer_input_gate",
|
| 265 |
+
"model.language_model.layers.32.per_layer_projection",
|
| 266 |
+
"model.language_model.layers.33.per_layer_input_gate",
|
| 267 |
+
"model.language_model.layers.33.per_layer_projection",
|
| 268 |
+
"model.language_model.layers.34.per_layer_input_gate",
|
| 269 |
+
"model.language_model.layers.34.per_layer_projection",
|
| 270 |
+
"model.language_model.layers.35.per_layer_input_gate",
|
| 271 |
+
"model.language_model.layers.35.per_layer_projection",
|
| 272 |
+
"model.language_model.layers.36.per_layer_input_gate",
|
| 273 |
+
"model.language_model.layers.36.per_layer_projection",
|
| 274 |
+
"model.language_model.layers.37.per_layer_input_gate",
|
| 275 |
+
"model.language_model.layers.37.per_layer_projection",
|
| 276 |
+
"model.language_model.layers.38.per_layer_input_gate",
|
| 277 |
+
"model.language_model.layers.38.per_layer_projection",
|
| 278 |
+
"model.language_model.layers.39.per_layer_input_gate",
|
| 279 |
+
"model.language_model.layers.39.per_layer_projection",
|
| 280 |
+
"model.language_model.layers.40.per_layer_input_gate",
|
| 281 |
+
"model.language_model.layers.40.per_layer_projection",
|
| 282 |
+
"model.language_model.layers.41.per_layer_input_gate",
|
| 283 |
+
"model.language_model.layers.41.per_layer_projection",
|
| 284 |
+
"model.audio_tower.subsample_conv_projection.input_proj_linear",
|
| 285 |
+
"model.audio_tower.layers.0.feed_forward1.ffw_layer_1.linear",
|
| 286 |
+
"model.audio_tower.layers.0.feed_forward1.ffw_layer_2.linear",
|
| 287 |
+
"model.audio_tower.layers.0.feed_forward2.ffw_layer_1.linear",
|
| 288 |
+
"model.audio_tower.layers.0.feed_forward2.ffw_layer_2.linear",
|
| 289 |
+
"model.audio_tower.layers.0.self_attn.q_proj.linear",
|
| 290 |
+
"model.audio_tower.layers.0.self_attn.k_proj.linear",
|
| 291 |
+
"model.audio_tower.layers.0.self_attn.v_proj.linear",
|
| 292 |
+
"model.audio_tower.layers.0.self_attn.post.linear",
|
| 293 |
+
"model.audio_tower.layers.0.self_attn.relative_k_proj",
|
| 294 |
+
"model.audio_tower.layers.0.lconv1d.linear_start.linear",
|
| 295 |
+
"model.audio_tower.layers.0.lconv1d.linear_end.linear",
|
| 296 |
+
"model.audio_tower.layers.1.feed_forward1.ffw_layer_1.linear",
|
| 297 |
+
"model.audio_tower.layers.1.feed_forward1.ffw_layer_2.linear",
|
| 298 |
+
"model.audio_tower.layers.1.feed_forward2.ffw_layer_1.linear",
|
| 299 |
+
"model.audio_tower.layers.1.feed_forward2.ffw_layer_2.linear",
|
| 300 |
+
"model.audio_tower.layers.1.self_attn.q_proj.linear",
|
| 301 |
+
"model.audio_tower.layers.1.self_attn.k_proj.linear",
|
| 302 |
+
"model.audio_tower.layers.1.self_attn.v_proj.linear",
|
| 303 |
+
"model.audio_tower.layers.1.self_attn.post.linear",
|
| 304 |
+
"model.audio_tower.layers.1.self_attn.relative_k_proj",
|
| 305 |
+
"model.audio_tower.layers.1.lconv1d.linear_start.linear",
|
| 306 |
+
"model.audio_tower.layers.1.lconv1d.linear_end.linear",
|
| 307 |
+
"model.audio_tower.layers.2.feed_forward1.ffw_layer_1.linear",
|
| 308 |
+
"model.audio_tower.layers.2.feed_forward1.ffw_layer_2.linear",
|
| 309 |
+
"model.audio_tower.layers.2.feed_forward2.ffw_layer_1.linear",
|
| 310 |
+
"model.audio_tower.layers.2.feed_forward2.ffw_layer_2.linear",
|
| 311 |
+
"model.audio_tower.layers.2.self_attn.q_proj.linear",
|
| 312 |
+
"model.audio_tower.layers.2.self_attn.k_proj.linear",
|
| 313 |
+
"model.audio_tower.layers.2.self_attn.v_proj.linear",
|
| 314 |
+
"model.audio_tower.layers.2.self_attn.post.linear",
|
| 315 |
+
"model.audio_tower.layers.2.self_attn.relative_k_proj",
|
| 316 |
+
"model.audio_tower.layers.2.lconv1d.linear_start.linear",
|
| 317 |
+
"model.audio_tower.layers.2.lconv1d.linear_end.linear",
|
| 318 |
+
"model.audio_tower.layers.3.feed_forward1.ffw_layer_1.linear",
|
| 319 |
+
"model.audio_tower.layers.3.feed_forward1.ffw_layer_2.linear",
|
| 320 |
+
"model.audio_tower.layers.3.feed_forward2.ffw_layer_1.linear",
|
| 321 |
+
"model.audio_tower.layers.3.feed_forward2.ffw_layer_2.linear",
|
| 322 |
+
"model.audio_tower.layers.3.self_attn.q_proj.linear",
|
| 323 |
+
"model.audio_tower.layers.3.self_attn.k_proj.linear",
|
| 324 |
+
"model.audio_tower.layers.3.self_attn.v_proj.linear",
|
| 325 |
+
"model.audio_tower.layers.3.self_attn.post.linear",
|
| 326 |
+
"model.audio_tower.layers.3.self_attn.relative_k_proj",
|
| 327 |
+
"model.audio_tower.layers.3.lconv1d.linear_start.linear",
|
| 328 |
+
"model.audio_tower.layers.3.lconv1d.linear_end.linear",
|
| 329 |
+
"model.audio_tower.layers.4.feed_forward1.ffw_layer_1.linear",
|
| 330 |
+
"model.audio_tower.layers.4.feed_forward1.ffw_layer_2.linear",
|
| 331 |
+
"model.audio_tower.layers.4.feed_forward2.ffw_layer_1.linear",
|
| 332 |
+
"model.audio_tower.layers.4.feed_forward2.ffw_layer_2.linear",
|
| 333 |
+
"model.audio_tower.layers.4.self_attn.q_proj.linear",
|
| 334 |
+
"model.audio_tower.layers.4.self_attn.k_proj.linear",
|
| 335 |
+
"model.audio_tower.layers.4.self_attn.v_proj.linear",
|
| 336 |
+
"model.audio_tower.layers.4.self_attn.post.linear",
|
| 337 |
+
"model.audio_tower.layers.4.self_attn.relative_k_proj",
|
| 338 |
+
"model.audio_tower.layers.4.lconv1d.linear_start.linear",
|
| 339 |
+
"model.audio_tower.layers.4.lconv1d.linear_end.linear",
|
| 340 |
+
"model.audio_tower.layers.5.feed_forward1.ffw_layer_1.linear",
|
| 341 |
+
"model.audio_tower.layers.5.feed_forward1.ffw_layer_2.linear",
|
| 342 |
+
"model.audio_tower.layers.5.feed_forward2.ffw_layer_1.linear",
|
| 343 |
+
"model.audio_tower.layers.5.feed_forward2.ffw_layer_2.linear",
|
| 344 |
+
"model.audio_tower.layers.5.self_attn.q_proj.linear",
|
| 345 |
+
"model.audio_tower.layers.5.self_attn.k_proj.linear",
|
| 346 |
+
"model.audio_tower.layers.5.self_attn.v_proj.linear",
|
| 347 |
+
"model.audio_tower.layers.5.self_attn.post.linear",
|
| 348 |
+
"model.audio_tower.layers.5.self_attn.relative_k_proj",
|
| 349 |
+
"model.audio_tower.layers.5.lconv1d.linear_start.linear",
|
| 350 |
+
"model.audio_tower.layers.5.lconv1d.linear_end.linear",
|
| 351 |
+
"model.audio_tower.layers.6.feed_forward1.ffw_layer_1.linear",
|
| 352 |
+
"model.audio_tower.layers.6.feed_forward1.ffw_layer_2.linear",
|
| 353 |
+
"model.audio_tower.layers.6.feed_forward2.ffw_layer_1.linear",
|
| 354 |
+
"model.audio_tower.layers.6.feed_forward2.ffw_layer_2.linear",
|
| 355 |
+
"model.audio_tower.layers.6.self_attn.q_proj.linear",
|
| 356 |
+
"model.audio_tower.layers.6.self_attn.k_proj.linear",
|
| 357 |
+
"model.audio_tower.layers.6.self_attn.v_proj.linear",
|
| 358 |
+
"model.audio_tower.layers.6.self_attn.post.linear",
|
| 359 |
+
"model.audio_tower.layers.6.self_attn.relative_k_proj",
|
| 360 |
+
"model.audio_tower.layers.6.lconv1d.linear_start.linear",
|
| 361 |
+
"model.audio_tower.layers.6.lconv1d.linear_end.linear",
|
| 362 |
+
"model.audio_tower.layers.7.feed_forward1.ffw_layer_1.linear",
|
| 363 |
+
"model.audio_tower.layers.7.feed_forward1.ffw_layer_2.linear",
|
| 364 |
+
"model.audio_tower.layers.7.feed_forward2.ffw_layer_1.linear",
|
| 365 |
+
"model.audio_tower.layers.7.feed_forward2.ffw_layer_2.linear",
|
| 366 |
+
"model.audio_tower.layers.7.self_attn.q_proj.linear",
|
| 367 |
+
"model.audio_tower.layers.7.self_attn.k_proj.linear",
|
| 368 |
+
"model.audio_tower.layers.7.self_attn.v_proj.linear",
|
| 369 |
+
"model.audio_tower.layers.7.self_attn.post.linear",
|
| 370 |
+
"model.audio_tower.layers.7.self_attn.relative_k_proj",
|
| 371 |
+
"model.audio_tower.layers.7.lconv1d.linear_start.linear",
|
| 372 |
+
"model.audio_tower.layers.7.lconv1d.linear_end.linear",
|
| 373 |
+
"model.audio_tower.layers.8.feed_forward1.ffw_layer_1.linear",
|
| 374 |
+
"model.audio_tower.layers.8.feed_forward1.ffw_layer_2.linear",
|
| 375 |
+
"model.audio_tower.layers.8.feed_forward2.ffw_layer_1.linear",
|
| 376 |
+
"model.audio_tower.layers.8.feed_forward2.ffw_layer_2.linear",
|
| 377 |
+
"model.audio_tower.layers.8.self_attn.q_proj.linear",
|
| 378 |
+
"model.audio_tower.layers.8.self_attn.k_proj.linear",
|
| 379 |
+
"model.audio_tower.layers.8.self_attn.v_proj.linear",
|
| 380 |
+
"model.audio_tower.layers.8.self_attn.post.linear",
|
| 381 |
+
"model.audio_tower.layers.8.self_attn.relative_k_proj",
|
| 382 |
+
"model.audio_tower.layers.8.lconv1d.linear_start.linear",
|
| 383 |
+
"model.audio_tower.layers.8.lconv1d.linear_end.linear",
|
| 384 |
+
"model.audio_tower.layers.9.feed_forward1.ffw_layer_1.linear",
|
| 385 |
+
"model.audio_tower.layers.9.feed_forward1.ffw_layer_2.linear",
|
| 386 |
+
"model.audio_tower.layers.9.feed_forward2.ffw_layer_1.linear",
|
| 387 |
+
"model.audio_tower.layers.9.feed_forward2.ffw_layer_2.linear",
|
| 388 |
+
"model.audio_tower.layers.9.self_attn.q_proj.linear",
|
| 389 |
+
"model.audio_tower.layers.9.self_attn.k_proj.linear",
|
| 390 |
+
"model.audio_tower.layers.9.self_attn.v_proj.linear",
|
| 391 |
+
"model.audio_tower.layers.9.self_attn.post.linear",
|
| 392 |
+
"model.audio_tower.layers.9.self_attn.relative_k_proj",
|
| 393 |
+
"model.audio_tower.layers.9.lconv1d.linear_start.linear",
|
| 394 |
+
"model.audio_tower.layers.9.lconv1d.linear_end.linear",
|
| 395 |
+
"model.audio_tower.layers.10.feed_forward1.ffw_layer_1.linear",
|
| 396 |
+
"model.audio_tower.layers.10.feed_forward1.ffw_layer_2.linear",
|
| 397 |
+
"model.audio_tower.layers.10.feed_forward2.ffw_layer_1.linear",
|
| 398 |
+
"model.audio_tower.layers.10.feed_forward2.ffw_layer_2.linear",
|
| 399 |
+
"model.audio_tower.layers.10.self_attn.q_proj.linear",
|
| 400 |
+
"model.audio_tower.layers.10.self_attn.k_proj.linear",
|
| 401 |
+
"model.audio_tower.layers.10.self_attn.v_proj.linear",
|
| 402 |
+
"model.audio_tower.layers.10.self_attn.post.linear",
|
| 403 |
+
"model.audio_tower.layers.10.self_attn.relative_k_proj",
|
| 404 |
+
"model.audio_tower.layers.10.lconv1d.linear_start.linear",
|
| 405 |
+
"model.audio_tower.layers.10.lconv1d.linear_end.linear",
|
| 406 |
+
"model.audio_tower.layers.11.feed_forward1.ffw_layer_1.linear",
|
| 407 |
+
"model.audio_tower.layers.11.feed_forward1.ffw_layer_2.linear",
|
| 408 |
+
"model.audio_tower.layers.11.feed_forward2.ffw_layer_1.linear",
|
| 409 |
+
"model.audio_tower.layers.11.feed_forward2.ffw_layer_2.linear",
|
| 410 |
+
"model.audio_tower.layers.11.self_attn.q_proj.linear",
|
| 411 |
+
"model.audio_tower.layers.11.self_attn.k_proj.linear",
|
| 412 |
+
"model.audio_tower.layers.11.self_attn.v_proj.linear",
|
| 413 |
+
"model.audio_tower.layers.11.self_attn.post.linear",
|
| 414 |
+
"model.audio_tower.layers.11.self_attn.relative_k_proj",
|
| 415 |
+
"model.audio_tower.layers.11.lconv1d.linear_start.linear",
|
| 416 |
+
"model.audio_tower.layers.11.lconv1d.linear_end.linear",
|
| 417 |
+
"model.audio_tower.output_proj",
|
| 418 |
+
"model.embed_vision.embedding_projection",
|
| 419 |
+
"model.embed_audio.embedding_projection",
|
| 420 |
+
"lm_head"
|
| 421 |
+
],
|
| 422 |
+
"kv_cache_scheme": null,
|
| 423 |
+
"quant_method": "compressed-tensors",
|
| 424 |
+
"quantization_status": "compressed",
|
| 425 |
+
"sparsity_config": {},
|
| 426 |
+
"transform_config": {},
|
| 427 |
+
"version": "0.1.dev483+ga551158"
|
| 428 |
+
},
|
| 429 |
"text_config": {
|
| 430 |
"attention_bias": false,
|
| 431 |
"attention_dropout": 0.0,
|
| 432 |
"attention_k_eq_v": false,
|
| 433 |
"bos_token_id": 2,
|
| 434 |
+
"dtype": "float16",
|
| 435 |
"enable_moe_block": false,
|
| 436 |
"eos_token_id": 1,
|
| 437 |
"expert_intermediate_size": null,
|
|
|
|
| 489 |
],
|
| 490 |
"max_position_embeddings": 131072,
|
| 491 |
"model_type": "gemma4_text",
|
| 492 |
+
"moe_intermediate_size": null,
|
| 493 |
"num_attention_heads": 8,
|
| 494 |
"num_experts": null,
|
| 495 |
"num_global_key_value_heads": null,
|
|
|
|
| 519 |
"vocab_size_per_layer_input": 262144
|
| 520 |
},
|
| 521 |
"tie_word_embeddings": true,
|
| 522 |
+
"transformers_version": "5.8.0.dev0",
|
| 523 |
"video_token_id": 258884,
|
| 524 |
"vision_config": {
|
| 525 |
"_name_or_path": "",
|
|
|
|
| 528 |
"attention_dropout": 0.0,
|
| 529 |
"chunk_size_feed_forward": 0,
|
| 530 |
"default_output_length": 280,
|
| 531 |
+
"dtype": "float16",
|
| 532 |
"global_head_dim": 64,
|
| 533 |
"head_dim": 64,
|
| 534 |
"hidden_activation": "gelu_pytorch_tanh",
|
|
|
|
| 565 |
"use_clipped_linears": true
|
| 566 |
},
|
| 567 |
"vision_soft_tokens_per_image": 280
|
| 568 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1f8dc375e759ad817f79df6c413679e1c97b07a6ff0c0a36b37778aaae1056d
|
| 3 |
+
size 10312042292
|
processor_config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"audio_ms_per_token":
|
| 3 |
"audio_seq_length": 750,
|
| 4 |
"feature_extractor": {
|
| 5 |
"dither": 0.0,
|
|
|
|
| 1 |
{
|
| 2 |
+
"audio_ms_per_token": 40,
|
| 3 |
"audio_seq_length": 750,
|
| 4 |
"feature_extractor": {
|
| 5 |
"dither": 0.0,
|