Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- config.json +41 -0
- mergekit_config.yml +738 -0
- model.safetensors +3 -0
- preprocessor_config.json +27 -0
- processor_config.json +7 -0
- special_tokens_map.json +30 -0
- tokenizer.json +3 -0
- tokenizer_config.json +0 -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 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nintwentydo/Razorback-12B-v0.2",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"ignore_index": -100,
|
| 7 |
+
"image_seq_length": 1,
|
| 8 |
+
"image_token_index": 10,
|
| 9 |
+
"model_type": "llava",
|
| 10 |
+
"projector_hidden_act": "gelu",
|
| 11 |
+
"text_config": {
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"intermediate_size": 14336,
|
| 15 |
+
"is_composition": true,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"model_type": "mistral",
|
| 18 |
+
"num_hidden_layers": 40,
|
| 19 |
+
"num_attention_heads": 32,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"rms_norm_eps": 1e-05,
|
| 22 |
+
"rope_theta": 1000000000.0,
|
| 23 |
+
"sliding_window": null,
|
| 24 |
+
"vocab_size": 131072
|
| 25 |
+
},
|
| 26 |
+
"torch_dtype": "bfloat16",
|
| 27 |
+
"transformers_version": "4.47.1",
|
| 28 |
+
"vision_config": {
|
| 29 |
+
"head_dim": 64,
|
| 30 |
+
"num_attention_heads": 16,
|
| 31 |
+
"hidden_act": "silu",
|
| 32 |
+
"image_size": 1024,
|
| 33 |
+
"is_composition": true,
|
| 34 |
+
"model_type": "pixtral",
|
| 35 |
+
"patch_size": 16,
|
| 36 |
+
"rope_theta": 10000.0,
|
| 37 |
+
"tie_word_embeddings": false
|
| 38 |
+
},
|
| 39 |
+
"vision_feature_layer": -1,
|
| 40 |
+
"vision_feature_select_strategy": "full"
|
| 41 |
+
}
|
mergekit_config.yml
ADDED
|
@@ -0,0 +1,738 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
merge_method: task_arithmetic
|
| 2 |
+
dtype: bfloat16
|
| 3 |
+
base_model: /opt/dlami/nvme/models/unsloth_Pixtral-12B-2409
|
| 4 |
+
models:
|
| 5 |
+
- model: /opt/dlami/nvme/models/TheDrummer_UnslopNemo-12B-v3_llava
|
| 6 |
+
parameters:
|
| 7 |
+
normalize: false
|
| 8 |
+
weight:
|
| 9 |
+
- filter: language_model.lm_head.weight
|
| 10 |
+
value: 0.027243
|
| 11 |
+
- filter: language_model.model.embed_tokens.weight
|
| 12 |
+
value: 0.14717
|
| 13 |
+
- filter: language_model.model.layers.39.mlp.gate_proj.weight
|
| 14 |
+
value: 0.128203
|
| 15 |
+
- filter: language_model.model.layers.39.mlp.up_proj.weight
|
| 16 |
+
value: 0.123174
|
| 17 |
+
- filter: language_model.model.layers.39.mlp.down_proj.weight
|
| 18 |
+
value: 0.098388
|
| 19 |
+
- filter: language_model.model.layers.38.mlp.down_proj.weight
|
| 20 |
+
value: 0.125974
|
| 21 |
+
- filter: language_model.model.layers.38.mlp.up_proj.weight
|
| 22 |
+
value: 0.151487
|
| 23 |
+
- filter: language_model.model.layers.38.mlp.gate_proj.weight
|
| 24 |
+
value: 0.154605
|
| 25 |
+
- filter: language_model.model.layers.37.mlp.gate_proj.weight
|
| 26 |
+
value: 0.160597
|
| 27 |
+
- filter: language_model.model.layers.37.mlp.up_proj.weight
|
| 28 |
+
value: 0.160597
|
| 29 |
+
- filter: language_model.model.layers.37.mlp.down_proj.weight
|
| 30 |
+
value: 0.154017
|
| 31 |
+
- filter: language_model.model.layers.36.mlp.up_proj.weight
|
| 32 |
+
value: 0.166257
|
| 33 |
+
- filter: language_model.model.layers.36.mlp.gate_proj.weight
|
| 34 |
+
value: 0.165639
|
| 35 |
+
- filter: language_model.model.layers.36.mlp.down_proj.weight
|
| 36 |
+
value: 0.165014
|
| 37 |
+
- filter: language_model.model.layers.35.mlp.gate_proj.weight
|
| 38 |
+
value: 0.17496
|
| 39 |
+
- filter: language_model.model.layers.35.mlp.up_proj.weight
|
| 40 |
+
value: 0.170804
|
| 41 |
+
- filter: language_model.model.layers.35.mlp.down_proj.weight
|
| 42 |
+
value: 0.170191
|
| 43 |
+
- filter: language_model.model.layers.4.mlp.gate_proj.weight
|
| 44 |
+
value: 0.186816
|
| 45 |
+
- filter: language_model.model.layers.6.mlp.gate_proj.weight
|
| 46 |
+
value: 0.186816
|
| 47 |
+
- filter: language_model.model.norm.weight
|
| 48 |
+
value: 0.700371
|
| 49 |
+
- filter: language_model.model.layers.34.mlp.gate_proj.weight
|
| 50 |
+
value: 0.175869
|
| 51 |
+
- filter: language_model.model.layers.7.mlp.gate_proj.weight
|
| 52 |
+
value: 0.184212
|
| 53 |
+
- filter: language_model.model.layers.33.mlp.gate_proj.weight
|
| 54 |
+
value: 0.177382
|
| 55 |
+
- filter: language_model.model.layers.34.mlp.up_proj.weight
|
| 56 |
+
value: 0.171061
|
| 57 |
+
- filter: language_model.model.layers.5.mlp.gate_proj.weight
|
| 58 |
+
value: 0.187836
|
| 59 |
+
- filter: language_model.model.layers.8.mlp.gate_proj.weight
|
| 60 |
+
value: 0.177382
|
| 61 |
+
- filter: language_model.model.layers.11.mlp.gate_proj.weight
|
| 62 |
+
value: 0.180445
|
| 63 |
+
- filter: language_model.model.layers.33.mlp.up_proj.weight
|
| 64 |
+
value: 0.171948
|
| 65 |
+
- filter: language_model.model.layers.34.mlp.down_proj.weight
|
| 66 |
+
value: 0.174078
|
| 67 |
+
- filter: language_model.model.layers.9.mlp.gate_proj.weight
|
| 68 |
+
value: 0.176204
|
| 69 |
+
- filter: language_model.model.layers.10.mlp.gate_proj.weight
|
| 70 |
+
value: 0.175606
|
| 71 |
+
- filter: language_model.model.layers.3.mlp.gate_proj.weight
|
| 72 |
+
value: 0.186236
|
| 73 |
+
- filter: language_model.model.layers.32.mlp.gate_proj.weight
|
| 74 |
+
value: 0.184118
|
| 75 |
+
- filter: language_model.model.layers.12.mlp.gate_proj.weight
|
| 76 |
+
value: 0.183562
|
| 77 |
+
- filter: language_model.model.layers.13.mlp.gate_proj.weight
|
| 78 |
+
value: 0.180853
|
| 79 |
+
- filter: language_model.model.layers.31.mlp.gate_proj.weight
|
| 80 |
+
value: 0.189408
|
| 81 |
+
- filter: language_model.model.layers.32.mlp.up_proj.weight
|
| 82 |
+
value: 0.172229
|
| 83 |
+
- filter: language_model.model.layers.33.mlp.down_proj.weight
|
| 84 |
+
value: 0.176549
|
| 85 |
+
- filter: language_model.model.layers.14.mlp.gate_proj.weight
|
| 86 |
+
value: 0.175944
|
| 87 |
+
- filter: language_model.model.layers.30.mlp.gate_proj.weight
|
| 88 |
+
value: 0.194776
|
| 89 |
+
- filter: language_model.model.layers.29.mlp.gate_proj.weight
|
| 90 |
+
value: 0.194269
|
| 91 |
+
- filter: language_model.model.layers.31.mlp.up_proj.weight
|
| 92 |
+
value: 0.174714
|
| 93 |
+
- filter: language_model.model.layers.32.mlp.down_proj.weight
|
| 94 |
+
value: 0.179092
|
| 95 |
+
- filter: language_model.model.layers.15.mlp.gate_proj.weight
|
| 96 |
+
value: 0.174089
|
| 97 |
+
- filter: language_model.model.layers.21.mlp.gate_proj.weight
|
| 98 |
+
value: 0.187243
|
| 99 |
+
- filter: language_model.model.layers.22.mlp.gate_proj.weight
|
| 100 |
+
value: 0.189419
|
| 101 |
+
- filter: language_model.model.layers.27.mlp.gate_proj.weight
|
| 102 |
+
value: 0.195916
|
| 103 |
+
- filter: language_model.model.layers.28.mlp.gate_proj.weight
|
| 104 |
+
value: 0.195916
|
| 105 |
+
- filter: language_model.model.layers.16.mlp.gate_proj.weight
|
| 106 |
+
value: 0.173457
|
| 107 |
+
- filter: language_model.model.layers.2.mlp.gate_proj.weight
|
| 108 |
+
value: 0.191059
|
| 109 |
+
- filter: language_model.model.layers.20.mlp.gate_proj.weight
|
| 110 |
+
value: 0.184494
|
| 111 |
+
- filter: language_model.model.layers.26.mlp.gate_proj.weight
|
| 112 |
+
value: 0.19541
|
| 113 |
+
- filter: language_model.model.layers.30.mlp.up_proj.weight
|
| 114 |
+
value: 0.175673
|
| 115 |
+
- filter: language_model.model.layers.18.mlp.gate_proj.weight
|
| 116 |
+
value: 0.179492
|
| 117 |
+
- filter: language_model.model.layers.19.mlp.gate_proj.weight
|
| 118 |
+
value: 0.181707
|
| 119 |
+
- filter: language_model.model.layers.23.mlp.gate_proj.weight
|
| 120 |
+
value: 0.192713
|
| 121 |
+
- filter: language_model.model.layers.24.mlp.gate_proj.weight
|
| 122 |
+
value: 0.194899
|
| 123 |
+
- filter: language_model.model.layers.25.mlp.gate_proj.weight
|
| 124 |
+
value: 0.194899
|
| 125 |
+
- filter: language_model.model.layers.29.mlp.up_proj.weight
|
| 126 |
+
value: 0.177272
|
| 127 |
+
- filter: language_model.model.layers.31.mlp.down_proj.weight
|
| 128 |
+
value: 0.181707
|
| 129 |
+
- filter: language_model.model.layers.17.mlp.gate_proj.weight
|
| 130 |
+
value: 0.176651
|
| 131 |
+
- filter: language_model.model.layers.28.mlp.up_proj.weight
|
| 132 |
+
value: 0.176651
|
| 133 |
+
- filter: language_model.model.layers.26.mlp.up_proj.weight
|
| 134 |
+
value: 0.179905
|
| 135 |
+
- filter: language_model.model.layers.27.mlp.up_proj.weight
|
| 136 |
+
value: 0.179905
|
| 137 |
+
- filter: language_model.model.layers.30.mlp.down_proj.weight
|
| 138 |
+
value: 0.182155
|
| 139 |
+
- filter: language_model.model.layers.1.mlp.gate_proj.weight
|
| 140 |
+
value: 0.195027
|
| 141 |
+
- filter: language_model.model.layers.23.mlp.up_proj.weight
|
| 142 |
+
value: 0.186066
|
| 143 |
+
- filter: language_model.model.layers.24.mlp.up_proj.weight
|
| 144 |
+
value: 0.183812
|
| 145 |
+
- filter: language_model.model.layers.25.mlp.up_proj.weight
|
| 146 |
+
value: 0.183812
|
| 147 |
+
- filter: language_model.model.layers.15.mlp.up_proj.weight
|
| 148 |
+
value: 0.19674
|
| 149 |
+
- filter: language_model.model.layers.16.mlp.up_proj.weight
|
| 150 |
+
value: 0.198974
|
| 151 |
+
- filter: language_model.model.layers.17.mlp.up_proj.weight
|
| 152 |
+
value: 0.201202
|
| 153 |
+
- filter: language_model.model.layers.21.mlp.up_proj.weight
|
| 154 |
+
value: 0.187747
|
| 155 |
+
- filter: language_model.model.layers.22.mlp.up_proj.weight
|
| 156 |
+
value: 0.185485
|
| 157 |
+
- filter: language_model.model.layers.0.mlp.gate_proj.weight
|
| 158 |
+
value: 0.191707
|
| 159 |
+
- filter: language_model.model.layers.14.mlp.up_proj.weight
|
| 160 |
+
value: 0.193966
|
| 161 |
+
- filter: language_model.model.layers.18.mlp.up_proj.weight
|
| 162 |
+
value: 0.200708
|
| 163 |
+
- filter: language_model.model.layers.19.mlp.up_proj.weight
|
| 164 |
+
value: 0.19622
|
| 165 |
+
- filter: language_model.model.layers.20.mlp.up_proj.weight
|
| 166 |
+
value: 0.191707
|
| 167 |
+
- filter: language_model.model.layers.29.mlp.down_proj.weight
|
| 168 |
+
value: 0.187172
|
| 169 |
+
- filter: language_model.model.layers.10.mlp.up_proj.weight
|
| 170 |
+
value: 0.195161
|
| 171 |
+
- filter: language_model.model.layers.9.mlp.up_proj.weight
|
| 172 |
+
value: 0.192881
|
| 173 |
+
- filter: language_model.model.layers.12.mlp.up_proj.weight
|
| 174 |
+
value: 0.194623
|
| 175 |
+
- filter: language_model.model.layers.13.mlp.up_proj.weight
|
| 176 |
+
value: 0.194623
|
| 177 |
+
- filter: language_model.model.layers.28.mlp.down_proj.weight
|
| 178 |
+
value: 0.192328
|
| 179 |
+
- filter: language_model.model.layers.8.mlp.up_proj.weight
|
| 180 |
+
value: 0.187722
|
| 181 |
+
- filter: language_model.model.layers.5.mlp.up_proj.weight
|
| 182 |
+
value: 0.182478
|
| 183 |
+
- filter: language_model.model.layers.11.mlp.up_proj.weight
|
| 184 |
+
value: 0.195844
|
| 185 |
+
- filter: language_model.model.layers.15.mlp.down_proj.weight
|
| 186 |
+
value: 0.202757
|
| 187 |
+
- filter: language_model.model.layers.27.mlp.down_proj.weight
|
| 188 |
+
value: 0.193527
|
| 189 |
+
- filter: language_model.model.layers.3.mlp.up_proj.weight
|
| 190 |
+
value: 0.179504
|
| 191 |
+
- filter: language_model.model.layers.14.mlp.down_proj.weight
|
| 192 |
+
value: 0.199945
|
| 193 |
+
- filter: language_model.model.layers.16.mlp.down_proj.weight
|
| 194 |
+
value: 0.206862
|
| 195 |
+
- filter: language_model.model.layers.2.mlp.up_proj.weight
|
| 196 |
+
value: 0.188289
|
| 197 |
+
- filter: language_model.model.layers.4.mlp.up_proj.weight
|
| 198 |
+
value: 0.181225
|
| 199 |
+
- filter: language_model.model.layers.6.mlp.up_proj.weight
|
| 200 |
+
value: 0.18594
|
| 201 |
+
- filter: language_model.model.layers.7.mlp.up_proj.weight
|
| 202 |
+
value: 0.190633
|
| 203 |
+
- filter: language_model.model.layers.26.mlp.down_proj.weight
|
| 204 |
+
value: 0.197091
|
| 205 |
+
- filter: language_model.model.layers.0.mlp.up_proj.weight
|
| 206 |
+
value: 0.191835
|
| 207 |
+
- filter: language_model.model.layers.8.mlp.down_proj.weight
|
| 208 |
+
value: 0.19655
|
| 209 |
+
- filter: language_model.model.layers.9.mlp.down_proj.weight
|
| 210 |
+
value: 0.198898
|
| 211 |
+
- filter: language_model.model.layers.1.mlp.up_proj.weight
|
| 212 |
+
value: 0.193633
|
| 213 |
+
- filter: language_model.model.layers.17.mlp.down_proj.weight
|
| 214 |
+
value: 0.212397
|
| 215 |
+
- filter: language_model.model.layers.10.mlp.down_proj.weight
|
| 216 |
+
value: 0.204918
|
| 217 |
+
- filter: language_model.model.layers.12.mlp.down_proj.weight
|
| 218 |
+
value: 0.202561
|
| 219 |
+
- filter: language_model.model.layers.13.mlp.down_proj.weight
|
| 220 |
+
value: 0.204918
|
| 221 |
+
- filter: language_model.model.layers.2.mlp.down_proj.weight
|
| 222 |
+
value: 0.195448
|
| 223 |
+
- filter: language_model.model.layers.21.mlp.down_proj.weight
|
| 224 |
+
value: 0.202561
|
| 225 |
+
- filter: language_model.model.layers.22.mlp.down_proj.weight
|
| 226 |
+
value: 0.202561
|
| 227 |
+
- filter: language_model.model.layers.25.mlp.down_proj.weight
|
| 228 |
+
value: 0.202561
|
| 229 |
+
- filter: language_model.model.layers.3.mlp.down_proj.weight
|
| 230 |
+
value: 0.190673
|
| 231 |
+
- filter: language_model.model.layers.23.mlp.down_proj.weight
|
| 232 |
+
value: 0.206783
|
| 233 |
+
- filter: language_model.model.layers.5.mlp.down_proj.weight
|
| 234 |
+
value: 0.194886
|
| 235 |
+
- filter: language_model.model.layers.18.mlp.down_proj.weight
|
| 236 |
+
value: 0.213388
|
| 237 |
+
- filter: language_model.model.layers.20.mlp.down_proj.weight
|
| 238 |
+
value: 0.208664
|
| 239 |
+
- filter: language_model.model.layers.24.mlp.down_proj.weight
|
| 240 |
+
value: 0.206291
|
| 241 |
+
- filter: language_model.model.layers.11.mlp.down_proj.weight
|
| 242 |
+
value: 0.205793
|
| 243 |
+
- filter: language_model.model.layers.19.mlp.down_proj.weight
|
| 244 |
+
value: 0.215302
|
| 245 |
+
- filter: language_model.model.layers.4.mlp.down_proj.weight
|
| 246 |
+
value: 0.193743
|
| 247 |
+
- filter: language_model.model.layers.6.mlp.down_proj.weight
|
| 248 |
+
value: 0.196167
|
| 249 |
+
- filter: language_model.model.layers.7.mlp.down_proj.weight
|
| 250 |
+
value: 0.200995
|
| 251 |
+
- filter: language_model.model.layers.1.mlp.down_proj.weight
|
| 252 |
+
value: 0.202878
|
| 253 |
+
- filter: language_model.model.layers.0.mlp.down_proj.weight
|
| 254 |
+
value: 0.216568
|
| 255 |
+
- filter: language_model.model.layers.39.self_attn.o_proj.weight
|
| 256 |
+
value: 0.125786
|
| 257 |
+
- filter: language_model.model.layers.38.self_attn.o_proj.weight
|
| 258 |
+
value: 0.168001
|
| 259 |
+
- filter: language_model.model.layers.39.self_attn.q_proj.weight
|
| 260 |
+
value: 0.104322
|
| 261 |
+
- filter: language_model.model.layers.32.self_attn.q_proj.weight
|
| 262 |
+
value: 0.142943
|
| 263 |
+
- filter: language_model.model.layers.35.self_attn.q_proj.weight
|
| 264 |
+
value: 0.135335
|
| 265 |
+
- filter: language_model.model.layers.27.self_attn.q_proj.weight
|
| 266 |
+
value: 0.155682
|
| 267 |
+
- filter: language_model.model.layers.28.self_attn.q_proj.weight
|
| 268 |
+
value: 0.160908
|
| 269 |
+
- filter: language_model.model.layers.36.self_attn.o_proj.weight
|
| 270 |
+
value: 0.169581
|
| 271 |
+
- filter: language_model.model.layers.30.self_attn.q_proj.weight
|
| 272 |
+
value: 0.153355
|
| 273 |
+
- filter: language_model.model.layers.33.self_attn.q_proj.weight
|
| 274 |
+
value: 0.14809
|
| 275 |
+
- filter: language_model.model.layers.37.self_attn.o_proj.weight
|
| 276 |
+
value: 0.181167
|
| 277 |
+
- filter: language_model.model.layers.29.self_attn.q_proj.weight
|
| 278 |
+
value: 0.151579
|
| 279 |
+
- filter: language_model.model.layers.35.self_attn.o_proj.weight
|
| 280 |
+
value: 0.172772
|
| 281 |
+
- filter: language_model.model.layers.5.self_attn.q_proj.weight
|
| 282 |
+
value: 0.20046
|
| 283 |
+
- filter: language_model.model.layers.7.self_attn.q_proj.weight
|
| 284 |
+
value: 0.19533
|
| 285 |
+
- filter: language_model.model.layers.36.self_attn.q_proj.weight
|
| 286 |
+
value: 0.131309
|
| 287 |
+
- filter: language_model.model.layers.6.self_attn.q_proj.weight
|
| 288 |
+
value: 0.208596
|
| 289 |
+
- filter: language_model.model.layers.25.self_attn.q_proj.weight
|
| 290 |
+
value: 0.169983
|
| 291 |
+
- filter: language_model.model.layers.31.self_attn.q_proj.weight
|
| 292 |
+
value: 0.159312
|
| 293 |
+
- filter: language_model.model.layers.2.self_attn.q_proj.weight
|
| 294 |
+
value: 0.226469
|
| 295 |
+
- filter: language_model.model.layers.22.self_attn.q_proj.weight
|
| 296 |
+
value: 0.173013
|
| 297 |
+
- filter: language_model.model.layers.26.self_attn.q_proj.weight
|
| 298 |
+
value: 0.169458
|
| 299 |
+
- filter: language_model.model.layers.34.self_attn.q_proj.weight
|
| 300 |
+
value: 0.138944
|
| 301 |
+
- filter: language_model.model.layers.38.self_attn.q_proj.weight
|
| 302 |
+
value: 0.120914
|
| 303 |
+
- filter: language_model.model.layers.4.self_attn.q_proj.weight
|
| 304 |
+
value: 0.206842
|
| 305 |
+
- filter: language_model.model.layers.3.self_attn.q_proj.weight
|
| 306 |
+
value: 0.218577
|
| 307 |
+
- filter: language_model.model.layers.24.self_attn.q_proj.weight
|
| 308 |
+
value: 0.17765
|
| 309 |
+
- filter: language_model.model.layers.13.self_attn.q_proj.weight
|
| 310 |
+
value: 0.19292
|
| 311 |
+
- filter: language_model.model.layers.32.self_attn.o_proj.weight
|
| 312 |
+
value: 0.167494
|
| 313 |
+
- filter: language_model.model.layers.8.self_attn.q_proj.weight
|
| 314 |
+
value: 0.203614
|
| 315 |
+
- filter: language_model.model.layers.9.self_attn.q_proj.weight
|
| 316 |
+
value: 0.201451
|
| 317 |
+
- filter: language_model.model.layers.33.self_attn.o_proj.weight
|
| 318 |
+
value: 0.162669
|
| 319 |
+
- filter: language_model.model.layers.34.self_attn.o_proj.weight
|
| 320 |
+
value: 0.190237
|
| 321 |
+
- filter: language_model.model.layers.21.self_attn.q_proj.weight
|
| 322 |
+
value: 0.191609
|
| 323 |
+
- filter: language_model.model.layers.12.self_attn.q_proj.weight
|
| 324 |
+
value: 0.196632
|
| 325 |
+
- filter: language_model.model.layers.23.self_attn.q_proj.weight
|
| 326 |
+
value: 0.180131
|
| 327 |
+
- filter: language_model.model.layers.11.self_attn.q_proj.weight
|
| 328 |
+
value: 0.207093
|
| 329 |
+
- filter: language_model.model.layers.37.self_attn.q_proj.weight
|
| 330 |
+
value: 0.132486
|
| 331 |
+
- filter: language_model.model.layers.10.self_attn.q_proj.weight
|
| 332 |
+
value: 0.209974
|
| 333 |
+
- filter: language_model.model.layers.30.self_attn.o_proj.weight
|
| 334 |
+
value: 0.159707
|
| 335 |
+
- filter: language_model.model.layers.14.self_attn.q_proj.weight
|
| 336 |
+
value: 0.202298
|
| 337 |
+
- filter: language_model.model.layers.15.self_attn.q_proj.weight
|
| 338 |
+
value: 0.204133
|
| 339 |
+
- filter: language_model.model.layers.20.self_attn.q_proj.weight
|
| 340 |
+
value: 0.196769
|
| 341 |
+
- filter: language_model.model.layers.18.self_attn.q_proj.weight
|
| 342 |
+
value: 0.212189
|
| 343 |
+
- filter: language_model.model.layers.19.self_attn.q_proj.weight
|
| 344 |
+
value: 0.196912
|
| 345 |
+
- filter: language_model.model.layers.1.self_attn.q_proj.weight
|
| 346 |
+
value: 0.257896
|
| 347 |
+
- filter: language_model.model.layers.16.self_attn.q_proj.weight
|
| 348 |
+
value: 0.212969
|
| 349 |
+
- filter: language_model.model.layers.29.self_attn.o_proj.weight
|
| 350 |
+
value: 0.174884
|
| 351 |
+
- filter: language_model.model.layers.31.self_attn.o_proj.weight
|
| 352 |
+
value: 0.17181
|
| 353 |
+
- filter: language_model.model.layers.17.self_attn.q_proj.weight
|
| 354 |
+
value: 0.211513
|
| 355 |
+
- filter: language_model.model.layers.28.self_attn.o_proj.weight
|
| 356 |
+
value: 0.166682
|
| 357 |
+
- filter: language_model.model.layers.27.self_attn.o_proj.weight
|
| 358 |
+
value: 0.168077
|
| 359 |
+
- filter: language_model.model.layers.26.self_attn.o_proj.weight
|
| 360 |
+
value: 0.173479
|
| 361 |
+
- filter: language_model.model.layers.0.self_attn.q_proj.weight
|
| 362 |
+
value: 0.257296
|
| 363 |
+
- filter: language_model.model.layers.25.self_attn.o_proj.weight
|
| 364 |
+
value: 0.186181
|
| 365 |
+
- filter: language_model.model.layers.22.self_attn.o_proj.weight
|
| 366 |
+
value: 0.198762
|
| 367 |
+
- filter: language_model.model.layers.24.self_attn.o_proj.weight
|
| 368 |
+
value: 0.196575
|
| 369 |
+
- filter: language_model.model.layers.23.self_attn.o_proj.weight
|
| 370 |
+
value: 0.196065
|
| 371 |
+
- filter: language_model.model.layers.21.self_attn.o_proj.weight
|
| 372 |
+
value: 0.204071
|
| 373 |
+
- filter: language_model.model.layers.2.self_attn.o_proj.weight
|
| 374 |
+
value: 0.172557
|
| 375 |
+
- filter: language_model.model.layers.3.self_attn.o_proj.weight
|
| 376 |
+
value: 0.158307
|
| 377 |
+
- filter: language_model.model.layers.4.self_attn.o_proj.weight
|
| 378 |
+
value: 0.181612
|
| 379 |
+
- filter: language_model.model.layers.20.self_attn.o_proj.weight
|
| 380 |
+
value: 0.21766
|
| 381 |
+
- filter: language_model.model.layers.5.self_attn.o_proj.weight
|
| 382 |
+
value: 0.17041
|
| 383 |
+
- filter: language_model.model.layers.19.self_attn.o_proj.weight
|
| 384 |
+
value: 0.233288
|
| 385 |
+
- filter: language_model.model.layers.0.self_attn.o_proj.weight
|
| 386 |
+
value: 0.126586
|
| 387 |
+
- filter: language_model.model.layers.6.self_attn.o_proj.weight
|
| 388 |
+
value: 0.187595
|
| 389 |
+
- filter: language_model.model.layers.7.self_attn.o_proj.weight
|
| 390 |
+
value: 0.20661
|
| 391 |
+
- filter: language_model.model.layers.1.self_attn.o_proj.weight
|
| 392 |
+
value: 0.17396
|
| 393 |
+
- filter: language_model.model.layers.14.self_attn.o_proj.weight
|
| 394 |
+
value: 0.216113
|
| 395 |
+
- filter: language_model.model.layers.8.self_attn.o_proj.weight
|
| 396 |
+
value: 0.21012
|
| 397 |
+
- filter: language_model.model.layers.12.self_attn.o_proj.weight
|
| 398 |
+
value: 0.239032
|
| 399 |
+
- filter: language_model.model.layers.18.self_attn.o_proj.weight
|
| 400 |
+
value: 0.239032
|
| 401 |
+
- filter: language_model.model.layers.9.self_attn.o_proj.weight
|
| 402 |
+
value: 0.234053
|
| 403 |
+
- filter: language_model.model.layers.13.self_attn.o_proj.weight
|
| 404 |
+
value: 0.235535
|
| 405 |
+
- filter: language_model.model.layers.11.self_attn.o_proj.weight
|
| 406 |
+
value: 0.235188
|
| 407 |
+
- filter: language_model.model.layers.15.self_attn.o_proj.weight
|
| 408 |
+
value: 0.240286
|
| 409 |
+
- filter: language_model.model.layers.38.self_attn.v_proj.weight
|
| 410 |
+
value: 0.245061
|
| 411 |
+
- filter: language_model.model.layers.17.self_attn.o_proj.weight
|
| 412 |
+
value: 0.259885
|
| 413 |
+
- filter: language_model.model.layers.39.self_attn.v_proj.weight
|
| 414 |
+
value: 0.174974
|
| 415 |
+
- filter: language_model.model.layers.10.self_attn.o_proj.weight
|
| 416 |
+
value: 0.24449
|
| 417 |
+
- filter: language_model.model.layers.16.self_attn.o_proj.weight
|
| 418 |
+
value: 0.247054
|
| 419 |
+
- filter: language_model.model.layers.0.self_attn.k_proj.weight
|
| 420 |
+
value: 0.263433
|
| 421 |
+
- filter: language_model.model.layers.37.self_attn.v_proj.weight
|
| 422 |
+
value: 0.251838
|
| 423 |
+
- filter: language_model.model.layers.36.self_attn.v_proj.weight
|
| 424 |
+
value: 0.229054
|
| 425 |
+
- filter: language_model.model.layers.2.self_attn.k_proj.weight
|
| 426 |
+
value: 0.272851
|
| 427 |
+
- filter: language_model.model.layers.34.self_attn.v_proj.weight
|
| 428 |
+
value: 0.237781
|
| 429 |
+
- filter: language_model.model.layers.35.self_attn.v_proj.weight
|
| 430 |
+
value: 0.234624
|
| 431 |
+
- filter: language_model.model.layers.39.self_attn.k_proj.weight
|
| 432 |
+
value: 0.131171
|
| 433 |
+
- filter: language_model.model.layers.4.self_attn.k_proj.weight
|
| 434 |
+
value: 0.283584
|
| 435 |
+
- filter: language_model.model.layers.6.self_attn.k_proj.weight
|
| 436 |
+
value: 0.27769
|
| 437 |
+
- filter: language_model.model.layers.5.self_attn.k_proj.weight
|
| 438 |
+
value: 0.271602
|
| 439 |
+
- filter: language_model.model.layers.7.self_attn.k_proj.weight
|
| 440 |
+
value: 0.262444
|
| 441 |
+
- filter: language_model.model.layers.3.self_attn.k_proj.weight
|
| 442 |
+
value: 0.287982
|
| 443 |
+
- filter: language_model.model.layers.1.self_attn.k_proj.weight
|
| 444 |
+
value: 0.322508
|
| 445 |
+
- filter: language_model.model.layers.8.self_attn.k_proj.weight
|
| 446 |
+
value: 0.27421
|
| 447 |
+
- filter: language_model.model.layers.13.self_attn.k_proj.weight
|
| 448 |
+
value: 0.264598
|
| 449 |
+
- filter: language_model.model.layers.9.self_attn.k_proj.weight
|
| 450 |
+
value: 0.270955
|
| 451 |
+
- filter: language_model.model.layers.28.self_attn.k_proj.weight
|
| 452 |
+
value: 0.225454
|
| 453 |
+
- filter: language_model.model.layers.32.self_attn.v_proj.weight
|
| 454 |
+
value: 0.223712
|
| 455 |
+
- filter: language_model.model.layers.10.self_attn.k_proj.weight
|
| 456 |
+
value: 0.280152
|
| 457 |
+
- filter: language_model.model.layers.11.self_attn.k_proj.weight
|
| 458 |
+
value: 0.281748
|
| 459 |
+
- filter: language_model.model.layers.12.self_attn.k_proj.weight
|
| 460 |
+
value: 0.268835
|
| 461 |
+
- filter: language_model.model.layers.15.self_attn.k_proj.weight
|
| 462 |
+
value: 0.278551
|
| 463 |
+
- filter: language_model.model.layers.30.self_attn.k_proj.weight
|
| 464 |
+
value: 0.224603
|
| 465 |
+
- filter: language_model.model.layers.25.self_attn.k_proj.weight
|
| 466 |
+
value: 0.246963
|
| 467 |
+
- filter: language_model.model.layers.27.self_attn.k_proj.weight
|
| 468 |
+
value: 0.233124
|
| 469 |
+
- filter: language_model.model.layers.35.self_attn.k_proj.weight
|
| 470 |
+
value: 0.189342
|
| 471 |
+
- filter: language_model.model.layers.14.self_attn.k_proj.weight
|
| 472 |
+
value: 0.275182
|
| 473 |
+
- filter: language_model.model.layers.26.self_attn.k_proj.weight
|
| 474 |
+
value: 0.239651
|
| 475 |
+
- filter: language_model.model.layers.29.self_attn.k_proj.weight
|
| 476 |
+
value: 0.212543
|
| 477 |
+
- filter: language_model.model.layers.31.self_attn.k_proj.weight
|
| 478 |
+
value: 0.224937
|
| 479 |
+
- filter: language_model.model.layers.22.self_attn.k_proj.weight
|
| 480 |
+
value: 0.247783
|
| 481 |
+
- filter: language_model.model.layers.32.self_attn.k_proj.weight
|
| 482 |
+
value: 0.200668
|
| 483 |
+
- filter: language_model.model.layers.33.self_attn.k_proj.weight
|
| 484 |
+
value: 0.215173
|
| 485 |
+
- filter: language_model.model.layers.24.self_attn.k_proj.weight
|
| 486 |
+
value: 0.24565
|
| 487 |
+
- filter: language_model.model.layers.29.self_attn.v_proj.weight
|
| 488 |
+
value: 0.220374
|
| 489 |
+
- filter: language_model.model.layers.33.self_attn.v_proj.weight
|
| 490 |
+
value: 0.227699
|
| 491 |
+
- filter: language_model.model.layers.18.self_attn.k_proj.weight
|
| 492 |
+
value: 0.2765
|
| 493 |
+
- filter: language_model.model.layers.36.self_attn.k_proj.weight
|
| 494 |
+
value: 0.17181
|
| 495 |
+
- filter: language_model.model.layers.16.self_attn.k_proj.weight
|
| 496 |
+
value: 0.289809
|
| 497 |
+
- filter: language_model.model.layers.21.self_attn.k_proj.weight
|
| 498 |
+
value: 0.266098
|
| 499 |
+
- filter: language_model.model.layers.23.self_attn.k_proj.weight
|
| 500 |
+
value: 0.246661
|
| 501 |
+
- filter: language_model.model.layers.20.self_attn.k_proj.weight
|
| 502 |
+
value: 0.266
|
| 503 |
+
- filter: language_model.model.layers.31.self_attn.v_proj.weight
|
| 504 |
+
value: 0.219631
|
| 505 |
+
- filter: language_model.model.layers.30.self_attn.v_proj.weight
|
| 506 |
+
value: 0.217368
|
| 507 |
+
- filter: language_model.model.layers.34.self_attn.k_proj.weight
|
| 508 |
+
value: 0.171745
|
| 509 |
+
- filter: language_model.model.layers.17.self_attn.k_proj.weight
|
| 510 |
+
value: 0.281311
|
| 511 |
+
- filter: language_model.model.layers.19.self_attn.k_proj.weight
|
| 512 |
+
value: 0.258162
|
| 513 |
+
- filter: language_model.model.layers.38.self_attn.k_proj.weight
|
| 514 |
+
value: 0.147819
|
| 515 |
+
- filter: language_model.model.layers.28.self_attn.v_proj.weight
|
| 516 |
+
value: 0.212052
|
| 517 |
+
- filter: language_model.model.layers.27.self_attn.v_proj.weight
|
| 518 |
+
value: 0.220392
|
| 519 |
+
- filter: language_model.model.layers.37.self_attn.k_proj.weight
|
| 520 |
+
value: 0.177511
|
| 521 |
+
- filter: language_model.model.layers.26.self_attn.v_proj.weight
|
| 522 |
+
value: 0.223488
|
| 523 |
+
- filter: language_model.model.layers.22.self_attn.v_proj.weight
|
| 524 |
+
value: 0.242903
|
| 525 |
+
- filter: language_model.model.layers.24.self_attn.v_proj.weight
|
| 526 |
+
value: 0.235792
|
| 527 |
+
- filter: language_model.model.layers.25.self_attn.v_proj.weight
|
| 528 |
+
value: 0.231352
|
| 529 |
+
- filter: language_model.model.layers.23.self_attn.v_proj.weight
|
| 530 |
+
value: 0.235188
|
| 531 |
+
- filter: language_model.model.layers.2.self_attn.v_proj.weight
|
| 532 |
+
value: 0.168169
|
| 533 |
+
- filter: language_model.model.layers.5.self_attn.v_proj.weight
|
| 534 |
+
value: 0.165651
|
| 535 |
+
- filter: language_model.model.layers.3.self_attn.v_proj.weight
|
| 536 |
+
value: 0.151622
|
| 537 |
+
- filter: language_model.model.layers.4.self_attn.v_proj.weight
|
| 538 |
+
value: 0.187611
|
| 539 |
+
- filter: language_model.model.layers.20.self_attn.v_proj.weight
|
| 540 |
+
value: 0.224346
|
| 541 |
+
- filter: language_model.model.layers.19.self_attn.v_proj.weight
|
| 542 |
+
value: 0.245325
|
| 543 |
+
- filter: language_model.model.layers.21.self_attn.v_proj.weight
|
| 544 |
+
value: 0.22807
|
| 545 |
+
- filter: language_model.model.layers.6.self_attn.v_proj.weight
|
| 546 |
+
value: 0.18168
|
| 547 |
+
- filter: language_model.model.layers.7.self_attn.v_proj.weight
|
| 548 |
+
value: 0.204672
|
| 549 |
+
- filter: language_model.model.layers.17.self_attn.v_proj.weight
|
| 550 |
+
value: 0.243401
|
| 551 |
+
- filter: language_model.model.layers.8.self_attn.v_proj.weight
|
| 552 |
+
value: 0.19022
|
| 553 |
+
- filter: language_model.model.layers.12.self_attn.v_proj.weight
|
| 554 |
+
value: 0.232988
|
| 555 |
+
- filter: language_model.model.layers.9.self_attn.v_proj.weight
|
| 556 |
+
value: 0.230431
|
| 557 |
+
- filter: language_model.model.layers.14.self_attn.v_proj.weight
|
| 558 |
+
value: 0.205856
|
| 559 |
+
- filter: language_model.model.layers.1.self_attn.v_proj.weight
|
| 560 |
+
value: 0.168527
|
| 561 |
+
- filter: language_model.model.layers.10.self_attn.v_proj.weight
|
| 562 |
+
value: 0.228984
|
| 563 |
+
- filter: language_model.model.layers.13.self_attn.v_proj.weight
|
| 564 |
+
value: 0.219502
|
| 565 |
+
- filter: language_model.model.layers.15.self_attn.v_proj.weight
|
| 566 |
+
value: 0.226354
|
| 567 |
+
- filter: language_model.model.layers.16.self_attn.v_proj.weight
|
| 568 |
+
value: 0.23457
|
| 569 |
+
- filter: language_model.model.layers.18.self_attn.v_proj.weight
|
| 570 |
+
value: 0.250678
|
| 571 |
+
- filter: language_model.model.layers.11.self_attn.v_proj.weight
|
| 572 |
+
value: 0.22923
|
| 573 |
+
- filter: language_model.model.layers.0.self_attn.v_proj.weight
|
| 574 |
+
value: 0.134132
|
| 575 |
+
- filter: language_model.model.layers.34.post_attention_layernorm.weight
|
| 576 |
+
value: 0.883075
|
| 577 |
+
- filter: language_model.model.layers.33.post_attention_layernorm.weight
|
| 578 |
+
value: 0.880867
|
| 579 |
+
- filter: language_model.model.layers.35.post_attention_layernorm.weight
|
| 580 |
+
value: 0.888537
|
| 581 |
+
- filter: language_model.model.layers.38.post_attention_layernorm.weight
|
| 582 |
+
value: 0.895727
|
| 583 |
+
- filter: language_model.model.layers.31.post_attention_layernorm.weight
|
| 584 |
+
value: 0.880202
|
| 585 |
+
- filter: language_model.model.layers.32.post_attention_layernorm.weight
|
| 586 |
+
value: 0.882647
|
| 587 |
+
- filter: language_model.model.layers.36.post_attention_layernorm.weight
|
| 588 |
+
value: 0.895084
|
| 589 |
+
- filter: language_model.model.layers.37.post_attention_layernorm.weight
|
| 590 |
+
value: 0.896463
|
| 591 |
+
- filter: language_model.model.layers.30.post_attention_layernorm.weight
|
| 592 |
+
value: 0.884693
|
| 593 |
+
- filter: language_model.model.layers.29.post_attention_layernorm.weight
|
| 594 |
+
value: 0.886562
|
| 595 |
+
- filter: language_model.model.layers.28.post_attention_layernorm.weight
|
| 596 |
+
value: 0.883506
|
| 597 |
+
- filter: language_model.model.layers.27.post_attention_layernorm.weight
|
| 598 |
+
value: 0.885518
|
| 599 |
+
- filter: language_model.model.layers.30.input_layernorm.weight
|
| 600 |
+
value: 0.900142
|
| 601 |
+
- filter: language_model.model.layers.26.post_attention_layernorm.weight
|
| 602 |
+
value: 0.888298
|
| 603 |
+
- filter: language_model.model.layers.38.input_layernorm.weight
|
| 604 |
+
value: 0.918705
|
| 605 |
+
- filter: language_model.model.layers.25.post_attention_layernorm.weight
|
| 606 |
+
value: 0.887788
|
| 607 |
+
- filter: language_model.model.layers.11.post_attention_layernorm.weight
|
| 608 |
+
value: 0.819136
|
| 609 |
+
- filter: language_model.model.layers.8.input_layernorm.weight
|
| 610 |
+
value: 0.827776
|
| 611 |
+
- filter: language_model.model.layers.12.post_attention_layernorm.weight
|
| 612 |
+
value: 0.829306
|
| 613 |
+
- filter: language_model.model.layers.17.post_attention_layernorm.weight
|
| 614 |
+
value: 0.844334
|
| 615 |
+
- filter: language_model.model.layers.24.post_attention_layernorm.weight
|
| 616 |
+
value: 0.88749
|
| 617 |
+
- filter: language_model.model.layers.23.post_attention_layernorm.weight
|
| 618 |
+
value: 0.882598
|
| 619 |
+
- filter: language_model.model.layers.18.post_attention_layernorm.weight
|
| 620 |
+
value: 0.855345
|
| 621 |
+
- filter: language_model.model.layers.16.post_attention_layernorm.weight
|
| 622 |
+
value: 0.842616
|
| 623 |
+
- filter: language_model.model.layers.27.input_layernorm.weight
|
| 624 |
+
value: 0.906137
|
| 625 |
+
- filter: language_model.model.layers.13.post_attention_layernorm.weight
|
| 626 |
+
value: 0.838716
|
| 627 |
+
- filter: language_model.model.layers.26.input_layernorm.weight
|
| 628 |
+
value: 0.901367
|
| 629 |
+
- filter: language_model.model.layers.10.post_attention_layernorm.weight
|
| 630 |
+
value: 0.815316
|
| 631 |
+
- filter: language_model.model.layers.14.input_layernorm.weight
|
| 632 |
+
value: 0.879509
|
| 633 |
+
- filter: language_model.model.layers.15.post_attention_layernorm.weight
|
| 634 |
+
value: 0.843714
|
| 635 |
+
- filter: language_model.model.layers.19.post_attention_layernorm.weight
|
| 636 |
+
value: 0.865915
|
| 637 |
+
- filter: language_model.model.layers.22.post_attention_layernorm.weight
|
| 638 |
+
value: 0.885566
|
| 639 |
+
- filter: language_model.model.layers.28.input_layernorm.weight
|
| 640 |
+
value: 0.902543
|
| 641 |
+
- filter: language_model.model.layers.5.input_layernorm.weight
|
| 642 |
+
value: 0.810565
|
| 643 |
+
- filter: language_model.model.layers.6.post_attention_layernorm.weight
|
| 644 |
+
value: 0.796305
|
| 645 |
+
- filter: language_model.model.layers.7.post_attention_layernorm.weight
|
| 646 |
+
value: 0.806038
|
| 647 |
+
- filter: language_model.model.layers.9.post_attention_layernorm.weight
|
| 648 |
+
value: 0.815692
|
| 649 |
+
- filter: language_model.model.layers.21.input_layernorm.weight
|
| 650 |
+
value: 0.891671
|
| 651 |
+
- filter: language_model.model.layers.6.input_layernorm.weight
|
| 652 |
+
value: 0.811936
|
| 653 |
+
- filter: language_model.model.layers.7.input_layernorm.weight
|
| 654 |
+
value: 0.83771
|
| 655 |
+
- filter: language_model.model.layers.14.post_attention_layernorm.weight
|
| 656 |
+
value: 0.852636
|
| 657 |
+
- filter: language_model.model.layers.21.post_attention_layernorm.weight
|
| 658 |
+
value: 0.886217
|
| 659 |
+
- filter: language_model.model.layers.20.post_attention_layernorm.weight
|
| 660 |
+
value: 0.880705
|
| 661 |
+
- filter: language_model.model.layers.3.input_layernorm.weight
|
| 662 |
+
value: 0.816342
|
| 663 |
+
- filter: language_model.model.layers.10.input_layernorm.weight
|
| 664 |
+
value: 0.863017
|
| 665 |
+
- filter: language_model.model.layers.8.post_attention_layernorm.weight
|
| 666 |
+
value: 0.820931
|
| 667 |
+
- filter: language_model.model.layers.1.input_layernorm.weight
|
| 668 |
+
value: 0.771411
|
| 669 |
+
- filter: language_model.model.layers.31.input_layernorm.weight
|
| 670 |
+
value: 0.910857
|
| 671 |
+
- filter: language_model.model.layers.37.input_layernorm.weight
|
| 672 |
+
value: 0.922558
|
| 673 |
+
- filter: language_model.model.layers.5.post_attention_layernorm.weight
|
| 674 |
+
value: 0.794374
|
| 675 |
+
- filter: language_model.model.layers.25.input_layernorm.weight
|
| 676 |
+
value: 0.9114
|
| 677 |
+
- filter: language_model.model.layers.35.input_layernorm.weight
|
| 678 |
+
value: 0.923684
|
| 679 |
+
- filter: language_model.model.layers.33.input_layernorm.weight
|
| 680 |
+
value: 0.919435
|
| 681 |
+
- filter: language_model.model.layers.16.input_layernorm.weight
|
| 682 |
+
value: 0.892982
|
| 683 |
+
- filter: language_model.model.layers.32.input_layernorm.weight
|
| 684 |
+
value: 0.921119
|
| 685 |
+
- filter: language_model.model.layers.4.post_attention_layernorm.weight
|
| 686 |
+
value: 0.784228
|
| 687 |
+
- filter: language_model.model.layers.9.input_layernorm.weight
|
| 688 |
+
value: 0.868681
|
| 689 |
+
- filter: language_model.model.layers.11.input_layernorm.weight
|
| 690 |
+
value: 0.884768
|
| 691 |
+
- filter: language_model.model.layers.36.input_layernorm.weight
|
| 692 |
+
value: 0.927183
|
| 693 |
+
- filter: language_model.model.layers.20.input_layernorm.weight
|
| 694 |
+
value: 0.912409
|
| 695 |
+
- filter: language_model.model.layers.29.input_layernorm.weight
|
| 696 |
+
value: 0.92094
|
| 697 |
+
- filter: language_model.model.layers.34.input_layernorm.weight
|
| 698 |
+
value: 0.924849
|
| 699 |
+
- filter: language_model.model.layers.15.input_layernorm.weight
|
| 700 |
+
value: 0.896905
|
| 701 |
+
- filter: language_model.model.layers.12.input_layernorm.weight
|
| 702 |
+
value: 0.891217
|
| 703 |
+
- filter: language_model.model.layers.13.input_layernorm.weight
|
| 704 |
+
value: 0.8992
|
| 705 |
+
- filter: language_model.model.layers.23.input_layernorm.weight
|
| 706 |
+
value: 0.913613
|
| 707 |
+
- filter: language_model.model.layers.18.input_layernorm.weight
|
| 708 |
+
value: 0.906696
|
| 709 |
+
- filter: language_model.model.layers.24.input_layernorm.weight
|
| 710 |
+
value: 0.91877
|
| 711 |
+
- filter: language_model.model.layers.3.post_attention_layernorm.weight
|
| 712 |
+
value: 0.786238
|
| 713 |
+
- filter: language_model.model.layers.2.post_attention_layernorm.weight
|
| 714 |
+
value: 0.777384
|
| 715 |
+
- filter: language_model.model.layers.39.input_layernorm.weight
|
| 716 |
+
value: 0.942583
|
| 717 |
+
- filter: language_model.model.layers.4.input_layernorm.weight
|
| 718 |
+
value: 0.83397
|
| 719 |
+
- filter: language_model.model.layers.22.input_layernorm.weight
|
| 720 |
+
value: 0.916358
|
| 721 |
+
- filter: language_model.model.layers.17.input_layernorm.weight
|
| 722 |
+
value: 0.91935
|
| 723 |
+
- filter: language_model.model.layers.39.post_attention_layernorm.weight
|
| 724 |
+
value: 0.955214
|
| 725 |
+
- filter: language_model.model.layers.19.input_layernorm.weight
|
| 726 |
+
value: 0.932588
|
| 727 |
+
- filter: language_model.model.layers.0.post_attention_layernorm.weight
|
| 728 |
+
value: 0.709289
|
| 729 |
+
- filter: language_model.model.layers.1.post_attention_layernorm.weight
|
| 730 |
+
value: 0.750784
|
| 731 |
+
- filter: language_model.model.layers.2.input_layernorm.weight
|
| 732 |
+
value: 0.851994
|
| 733 |
+
- filter: language_model.model.layers.0.input_layernorm.weight
|
| 734 |
+
value: 0.736801
|
| 735 |
+
- value: 0.0
|
| 736 |
+
tokenizer:
|
| 737 |
+
source: base
|
| 738 |
+
chat_template: pixtral12b
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42bcd85f51bead20c5f7b8b86e1591e60ad86a296135eebcc34c651c4baef007
|
| 3 |
+
size 25365558112
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": true,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.48145466,
|
| 8 |
+
0.4578275,
|
| 9 |
+
0.40821073
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "PixtralImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.26862954,
|
| 14 |
+
0.26130258,
|
| 15 |
+
0.27577711
|
| 16 |
+
],
|
| 17 |
+
"patch_size": {
|
| 18 |
+
"height": 16,
|
| 19 |
+
"width": 16
|
| 20 |
+
},
|
| 21 |
+
"processor_class": "PixtralProcessor",
|
| 22 |
+
"resample": 3,
|
| 23 |
+
"rescale_factor": 0.00392156862745098,
|
| 24 |
+
"size": {
|
| 25 |
+
"longest_edge": 1024
|
| 26 |
+
}
|
| 27 |
+
}
|
processor_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_break_token": "[IMG_BREAK]",
|
| 3 |
+
"image_end_token": "[IMG_END]",
|
| 4 |
+
"image_token": "[IMG]",
|
| 5 |
+
"patch_size": 16,
|
| 6 |
+
"processor_class": "PixtralProcessor"
|
| 7 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<pad>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:84f33e6f52b2833e8cc17229af8eea363f640a898f19a48184a2c7f6f5a88337
|
| 3 |
+
size 17077329
|
tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|