sadkins65 commited on
Commit
b73df74
·
verified ·
1 Parent(s): c5a975c

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -27,8 +27,40 @@
27
  }
28
  },
29
  "format": "pack-quantized",
30
- "global_compression_ratio": 2.1823781700242217,
31
  "ignore": [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
32
  "lm_head"
33
  ],
34
  "kv_cache_scheme": null,
@@ -36,7 +68,7 @@
36
  "quantization_status": "frozen",
37
  "sparsity_config": {
38
  "format": "dense",
39
- "global_sparsity": 0.15880852433171147,
40
  "registry_requires_subclass": false,
41
  "sparsity_structure": "unstructured"
42
  }
 
27
  }
28
  },
29
  "format": "pack-quantized",
30
+ "global_compression_ratio": 2.0968847636762966,
31
  "ignore": [
32
+ "model.layers.0.block_sparse_moe.gate",
33
+ "model.layers.1.block_sparse_moe.gate",
34
+ "model.layers.2.block_sparse_moe.gate",
35
+ "model.layers.3.block_sparse_moe.gate",
36
+ "model.layers.4.block_sparse_moe.gate",
37
+ "model.layers.5.block_sparse_moe.gate",
38
+ "model.layers.6.block_sparse_moe.gate",
39
+ "model.layers.7.block_sparse_moe.gate",
40
+ "model.layers.8.block_sparse_moe.gate",
41
+ "model.layers.9.block_sparse_moe.gate",
42
+ "model.layers.10.block_sparse_moe.gate",
43
+ "model.layers.11.block_sparse_moe.gate",
44
+ "model.layers.12.block_sparse_moe.gate",
45
+ "model.layers.13.block_sparse_moe.gate",
46
+ "model.layers.14.block_sparse_moe.gate",
47
+ "model.layers.15.block_sparse_moe.gate",
48
+ "model.layers.16.block_sparse_moe.gate",
49
+ "model.layers.17.block_sparse_moe.gate",
50
+ "model.layers.18.block_sparse_moe.gate",
51
+ "model.layers.19.block_sparse_moe.gate",
52
+ "model.layers.20.block_sparse_moe.gate",
53
+ "model.layers.21.block_sparse_moe.gate",
54
+ "model.layers.22.block_sparse_moe.gate",
55
+ "model.layers.23.block_sparse_moe.gate",
56
+ "model.layers.24.block_sparse_moe.gate",
57
+ "model.layers.25.block_sparse_moe.gate",
58
+ "model.layers.26.block_sparse_moe.gate",
59
+ "model.layers.27.block_sparse_moe.gate",
60
+ "model.layers.28.block_sparse_moe.gate",
61
+ "model.layers.29.block_sparse_moe.gate",
62
+ "model.layers.30.block_sparse_moe.gate",
63
+ "model.layers.31.block_sparse_moe.gate",
64
  "lm_head"
65
  ],
66
  "kv_cache_scheme": null,
 
68
  "quantization_status": "frozen",
69
  "sparsity_config": {
70
  "format": "dense",
71
+ "global_sparsity": 0.1588017024690434,
72
  "registry_requires_subclass": false,
73
  "sparsity_structure": "unstructured"
74
  }
model-00001-of-00005.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:203a9eb11275ff428f84f80fde387fd9096b385a789061e46dc842f46033580d
3
- size 4985745808
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80356de2146cfe15890405bcbe8e2c69a26fea557a631ccb972d0c0d99862b76
3
+ size 4986084536
model-00002-of-00005.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:84c2deb8163905cf3b65d737eab333a3fc97c4b32b587630471b0bb841f7b110
3
- size 4974458544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a5aa1d5ee0fdf7a8df95b9dad3c95d6721a63f913ebb9fcda79be31f01479c2
3
+ size 4974748880
model-00003-of-00005.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d5ae426b1d41e617d05b489b9c152ee69631d1834bfce678620ad76f5a30caa7
3
- size 4996121080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1836d8ae0a57a331a2516cb82a052054dfbcecdeef5ee343fb78f44800de74e7
3
+ size 4996459792
model-00004-of-00005.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eda77f6ebce9f9b21a9367dc39f36be7a61e8daa2e9948350e66f3f3651a2062
3
- size 4996121080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80b4bc548dcf6df9f48beb2388238cee432cb3c15789782843c2bd4acf02b5d4
3
+ size 4996459792
model-00005-of-00005.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:48fccb390f8bbab17213c9680cfd1be9c72ad50aba4f44fbd960913b0564bd8a
3
- size 4518570688
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:660a49aed1b85e276e4767868bfdd9ed5ab2a47890de7f7564bbcd53b59ff111
3
+ size 4518812624
model.safetensors.index.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "metadata": {
3
- "total_size": 24470657536
4
  },
5
  "weight_map": {
6
  "lm_head.weight": "model-00005-of-00005.safetensors",
@@ -77,9 +77,7 @@
77
  "model.layers.0.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
78
  "model.layers.0.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
79
  "model.layers.0.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
80
- "model.layers.0.block_sparse_moe.gate.weight_packed": "model-00001-of-00005.safetensors",
81
- "model.layers.0.block_sparse_moe.gate.weight_scale": "model-00001-of-00005.safetensors",
82
- "model.layers.0.block_sparse_moe.gate.weight_shape": "model-00001-of-00005.safetensors",
83
  "model.layers.0.input_layernorm.weight": "model-00001-of-00005.safetensors",
84
  "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
85
  "model.layers.0.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
@@ -166,9 +164,7 @@
166
  "model.layers.1.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
167
  "model.layers.1.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
168
  "model.layers.1.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
169
- "model.layers.1.block_sparse_moe.gate.weight_packed": "model-00001-of-00005.safetensors",
170
- "model.layers.1.block_sparse_moe.gate.weight_scale": "model-00001-of-00005.safetensors",
171
- "model.layers.1.block_sparse_moe.gate.weight_shape": "model-00001-of-00005.safetensors",
172
  "model.layers.1.input_layernorm.weight": "model-00001-of-00005.safetensors",
173
  "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
174
  "model.layers.1.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
@@ -255,9 +251,7 @@
255
  "model.layers.10.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
256
  "model.layers.10.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
257
  "model.layers.10.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
258
- "model.layers.10.block_sparse_moe.gate.weight_packed": "model-00002-of-00005.safetensors",
259
- "model.layers.10.block_sparse_moe.gate.weight_scale": "model-00002-of-00005.safetensors",
260
- "model.layers.10.block_sparse_moe.gate.weight_shape": "model-00002-of-00005.safetensors",
261
  "model.layers.10.input_layernorm.weight": "model-00002-of-00005.safetensors",
262
  "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
263
  "model.layers.10.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
@@ -344,9 +338,7 @@
344
  "model.layers.11.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
345
  "model.layers.11.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
346
  "model.layers.11.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
347
- "model.layers.11.block_sparse_moe.gate.weight_packed": "model-00002-of-00005.safetensors",
348
- "model.layers.11.block_sparse_moe.gate.weight_scale": "model-00002-of-00005.safetensors",
349
- "model.layers.11.block_sparse_moe.gate.weight_shape": "model-00002-of-00005.safetensors",
350
  "model.layers.11.input_layernorm.weight": "model-00002-of-00005.safetensors",
351
  "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
352
  "model.layers.11.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
@@ -433,9 +425,7 @@
433
  "model.layers.12.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
434
  "model.layers.12.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
435
  "model.layers.12.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
436
- "model.layers.12.block_sparse_moe.gate.weight_packed": "model-00002-of-00005.safetensors",
437
- "model.layers.12.block_sparse_moe.gate.weight_scale": "model-00002-of-00005.safetensors",
438
- "model.layers.12.block_sparse_moe.gate.weight_shape": "model-00002-of-00005.safetensors",
439
  "model.layers.12.input_layernorm.weight": "model-00003-of-00005.safetensors",
440
  "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
441
  "model.layers.12.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
@@ -522,9 +512,7 @@
522
  "model.layers.13.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
523
  "model.layers.13.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
524
  "model.layers.13.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
525
- "model.layers.13.block_sparse_moe.gate.weight_packed": "model-00003-of-00005.safetensors",
526
- "model.layers.13.block_sparse_moe.gate.weight_scale": "model-00003-of-00005.safetensors",
527
- "model.layers.13.block_sparse_moe.gate.weight_shape": "model-00003-of-00005.safetensors",
528
  "model.layers.13.input_layernorm.weight": "model-00003-of-00005.safetensors",
529
  "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
530
  "model.layers.13.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
@@ -611,9 +599,7 @@
611
  "model.layers.14.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
612
  "model.layers.14.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
613
  "model.layers.14.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
614
- "model.layers.14.block_sparse_moe.gate.weight_packed": "model-00003-of-00005.safetensors",
615
- "model.layers.14.block_sparse_moe.gate.weight_scale": "model-00003-of-00005.safetensors",
616
- "model.layers.14.block_sparse_moe.gate.weight_shape": "model-00003-of-00005.safetensors",
617
  "model.layers.14.input_layernorm.weight": "model-00003-of-00005.safetensors",
618
  "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
619
  "model.layers.14.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
@@ -700,9 +686,7 @@
700
  "model.layers.15.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
701
  "model.layers.15.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
702
  "model.layers.15.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
703
- "model.layers.15.block_sparse_moe.gate.weight_packed": "model-00003-of-00005.safetensors",
704
- "model.layers.15.block_sparse_moe.gate.weight_scale": "model-00003-of-00005.safetensors",
705
- "model.layers.15.block_sparse_moe.gate.weight_shape": "model-00003-of-00005.safetensors",
706
  "model.layers.15.input_layernorm.weight": "model-00003-of-00005.safetensors",
707
  "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
708
  "model.layers.15.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
@@ -789,9 +773,7 @@
789
  "model.layers.16.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
790
  "model.layers.16.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
791
  "model.layers.16.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
792
- "model.layers.16.block_sparse_moe.gate.weight_packed": "model-00003-of-00005.safetensors",
793
- "model.layers.16.block_sparse_moe.gate.weight_scale": "model-00003-of-00005.safetensors",
794
- "model.layers.16.block_sparse_moe.gate.weight_shape": "model-00003-of-00005.safetensors",
795
  "model.layers.16.input_layernorm.weight": "model-00003-of-00005.safetensors",
796
  "model.layers.16.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
797
  "model.layers.16.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
@@ -878,9 +860,7 @@
878
  "model.layers.17.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
879
  "model.layers.17.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
880
  "model.layers.17.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
881
- "model.layers.17.block_sparse_moe.gate.weight_packed": "model-00003-of-00005.safetensors",
882
- "model.layers.17.block_sparse_moe.gate.weight_scale": "model-00003-of-00005.safetensors",
883
- "model.layers.17.block_sparse_moe.gate.weight_shape": "model-00003-of-00005.safetensors",
884
  "model.layers.17.input_layernorm.weight": "model-00003-of-00005.safetensors",
885
  "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
886
  "model.layers.17.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
@@ -967,9 +947,7 @@
967
  "model.layers.18.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
968
  "model.layers.18.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
969
  "model.layers.18.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
970
- "model.layers.18.block_sparse_moe.gate.weight_packed": "model-00003-of-00005.safetensors",
971
- "model.layers.18.block_sparse_moe.gate.weight_scale": "model-00003-of-00005.safetensors",
972
- "model.layers.18.block_sparse_moe.gate.weight_shape": "model-00003-of-00005.safetensors",
973
  "model.layers.18.input_layernorm.weight": "model-00003-of-00005.safetensors",
974
  "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
975
  "model.layers.18.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
@@ -1056,9 +1034,7 @@
1056
  "model.layers.19.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1057
  "model.layers.19.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1058
  "model.layers.19.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1059
- "model.layers.19.block_sparse_moe.gate.weight_packed": "model-00003-of-00005.safetensors",
1060
- "model.layers.19.block_sparse_moe.gate.weight_scale": "model-00003-of-00005.safetensors",
1061
- "model.layers.19.block_sparse_moe.gate.weight_shape": "model-00003-of-00005.safetensors",
1062
  "model.layers.19.input_layernorm.weight": "model-00004-of-00005.safetensors",
1063
  "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1064
  "model.layers.19.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
@@ -1145,9 +1121,7 @@
1145
  "model.layers.2.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
1146
  "model.layers.2.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
1147
  "model.layers.2.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
1148
- "model.layers.2.block_sparse_moe.gate.weight_packed": "model-00001-of-00005.safetensors",
1149
- "model.layers.2.block_sparse_moe.gate.weight_scale": "model-00001-of-00005.safetensors",
1150
- "model.layers.2.block_sparse_moe.gate.weight_shape": "model-00001-of-00005.safetensors",
1151
  "model.layers.2.input_layernorm.weight": "model-00001-of-00005.safetensors",
1152
  "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
1153
  "model.layers.2.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
@@ -1234,9 +1208,7 @@
1234
  "model.layers.20.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1235
  "model.layers.20.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1236
  "model.layers.20.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1237
- "model.layers.20.block_sparse_moe.gate.weight_packed": "model-00004-of-00005.safetensors",
1238
- "model.layers.20.block_sparse_moe.gate.weight_scale": "model-00004-of-00005.safetensors",
1239
- "model.layers.20.block_sparse_moe.gate.weight_shape": "model-00004-of-00005.safetensors",
1240
  "model.layers.20.input_layernorm.weight": "model-00004-of-00005.safetensors",
1241
  "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1242
  "model.layers.20.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
@@ -1323,9 +1295,7 @@
1323
  "model.layers.21.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1324
  "model.layers.21.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1325
  "model.layers.21.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1326
- "model.layers.21.block_sparse_moe.gate.weight_packed": "model-00004-of-00005.safetensors",
1327
- "model.layers.21.block_sparse_moe.gate.weight_scale": "model-00004-of-00005.safetensors",
1328
- "model.layers.21.block_sparse_moe.gate.weight_shape": "model-00004-of-00005.safetensors",
1329
  "model.layers.21.input_layernorm.weight": "model-00004-of-00005.safetensors",
1330
  "model.layers.21.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1331
  "model.layers.21.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
@@ -1412,9 +1382,7 @@
1412
  "model.layers.22.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1413
  "model.layers.22.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1414
  "model.layers.22.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1415
- "model.layers.22.block_sparse_moe.gate.weight_packed": "model-00004-of-00005.safetensors",
1416
- "model.layers.22.block_sparse_moe.gate.weight_scale": "model-00004-of-00005.safetensors",
1417
- "model.layers.22.block_sparse_moe.gate.weight_shape": "model-00004-of-00005.safetensors",
1418
  "model.layers.22.input_layernorm.weight": "model-00004-of-00005.safetensors",
1419
  "model.layers.22.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1420
  "model.layers.22.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
@@ -1501,9 +1469,7 @@
1501
  "model.layers.23.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1502
  "model.layers.23.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1503
  "model.layers.23.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1504
- "model.layers.23.block_sparse_moe.gate.weight_packed": "model-00004-of-00005.safetensors",
1505
- "model.layers.23.block_sparse_moe.gate.weight_scale": "model-00004-of-00005.safetensors",
1506
- "model.layers.23.block_sparse_moe.gate.weight_shape": "model-00004-of-00005.safetensors",
1507
  "model.layers.23.input_layernorm.weight": "model-00004-of-00005.safetensors",
1508
  "model.layers.23.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1509
  "model.layers.23.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
@@ -1590,9 +1556,7 @@
1590
  "model.layers.24.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1591
  "model.layers.24.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1592
  "model.layers.24.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1593
- "model.layers.24.block_sparse_moe.gate.weight_packed": "model-00004-of-00005.safetensors",
1594
- "model.layers.24.block_sparse_moe.gate.weight_scale": "model-00004-of-00005.safetensors",
1595
- "model.layers.24.block_sparse_moe.gate.weight_shape": "model-00004-of-00005.safetensors",
1596
  "model.layers.24.input_layernorm.weight": "model-00004-of-00005.safetensors",
1597
  "model.layers.24.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1598
  "model.layers.24.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
@@ -1679,9 +1643,7 @@
1679
  "model.layers.25.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1680
  "model.layers.25.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1681
  "model.layers.25.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1682
- "model.layers.25.block_sparse_moe.gate.weight_packed": "model-00004-of-00005.safetensors",
1683
- "model.layers.25.block_sparse_moe.gate.weight_scale": "model-00004-of-00005.safetensors",
1684
- "model.layers.25.block_sparse_moe.gate.weight_shape": "model-00004-of-00005.safetensors",
1685
  "model.layers.25.input_layernorm.weight": "model-00004-of-00005.safetensors",
1686
  "model.layers.25.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1687
  "model.layers.25.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
@@ -1768,9 +1730,7 @@
1768
  "model.layers.26.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
1769
  "model.layers.26.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
1770
  "model.layers.26.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
1771
- "model.layers.26.block_sparse_moe.gate.weight_packed": "model-00004-of-00005.safetensors",
1772
- "model.layers.26.block_sparse_moe.gate.weight_scale": "model-00004-of-00005.safetensors",
1773
- "model.layers.26.block_sparse_moe.gate.weight_shape": "model-00004-of-00005.safetensors",
1774
  "model.layers.26.input_layernorm.weight": "model-00005-of-00005.safetensors",
1775
  "model.layers.26.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
1776
  "model.layers.26.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
@@ -1857,9 +1817,7 @@
1857
  "model.layers.27.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
1858
  "model.layers.27.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
1859
  "model.layers.27.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
1860
- "model.layers.27.block_sparse_moe.gate.weight_packed": "model-00005-of-00005.safetensors",
1861
- "model.layers.27.block_sparse_moe.gate.weight_scale": "model-00005-of-00005.safetensors",
1862
- "model.layers.27.block_sparse_moe.gate.weight_shape": "model-00005-of-00005.safetensors",
1863
  "model.layers.27.input_layernorm.weight": "model-00005-of-00005.safetensors",
1864
  "model.layers.27.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
1865
  "model.layers.27.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
@@ -1946,9 +1904,7 @@
1946
  "model.layers.28.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
1947
  "model.layers.28.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
1948
  "model.layers.28.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
1949
- "model.layers.28.block_sparse_moe.gate.weight_packed": "model-00005-of-00005.safetensors",
1950
- "model.layers.28.block_sparse_moe.gate.weight_scale": "model-00005-of-00005.safetensors",
1951
- "model.layers.28.block_sparse_moe.gate.weight_shape": "model-00005-of-00005.safetensors",
1952
  "model.layers.28.input_layernorm.weight": "model-00005-of-00005.safetensors",
1953
  "model.layers.28.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
1954
  "model.layers.28.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
@@ -2035,9 +1991,7 @@
2035
  "model.layers.29.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
2036
  "model.layers.29.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
2037
  "model.layers.29.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
2038
- "model.layers.29.block_sparse_moe.gate.weight_packed": "model-00005-of-00005.safetensors",
2039
- "model.layers.29.block_sparse_moe.gate.weight_scale": "model-00005-of-00005.safetensors",
2040
- "model.layers.29.block_sparse_moe.gate.weight_shape": "model-00005-of-00005.safetensors",
2041
  "model.layers.29.input_layernorm.weight": "model-00005-of-00005.safetensors",
2042
  "model.layers.29.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
2043
  "model.layers.29.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
@@ -2124,9 +2078,7 @@
2124
  "model.layers.3.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
2125
  "model.layers.3.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
2126
  "model.layers.3.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
2127
- "model.layers.3.block_sparse_moe.gate.weight_packed": "model-00001-of-00005.safetensors",
2128
- "model.layers.3.block_sparse_moe.gate.weight_scale": "model-00001-of-00005.safetensors",
2129
- "model.layers.3.block_sparse_moe.gate.weight_shape": "model-00001-of-00005.safetensors",
2130
  "model.layers.3.input_layernorm.weight": "model-00001-of-00005.safetensors",
2131
  "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
2132
  "model.layers.3.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
@@ -2213,9 +2165,7 @@
2213
  "model.layers.30.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
2214
  "model.layers.30.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
2215
  "model.layers.30.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
2216
- "model.layers.30.block_sparse_moe.gate.weight_packed": "model-00005-of-00005.safetensors",
2217
- "model.layers.30.block_sparse_moe.gate.weight_scale": "model-00005-of-00005.safetensors",
2218
- "model.layers.30.block_sparse_moe.gate.weight_shape": "model-00005-of-00005.safetensors",
2219
  "model.layers.30.input_layernorm.weight": "model-00005-of-00005.safetensors",
2220
  "model.layers.30.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
2221
  "model.layers.30.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
@@ -2302,9 +2252,7 @@
2302
  "model.layers.31.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
2303
  "model.layers.31.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
2304
  "model.layers.31.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
2305
- "model.layers.31.block_sparse_moe.gate.weight_packed": "model-00005-of-00005.safetensors",
2306
- "model.layers.31.block_sparse_moe.gate.weight_scale": "model-00005-of-00005.safetensors",
2307
- "model.layers.31.block_sparse_moe.gate.weight_shape": "model-00005-of-00005.safetensors",
2308
  "model.layers.31.input_layernorm.weight": "model-00005-of-00005.safetensors",
2309
  "model.layers.31.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
2310
  "model.layers.31.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
@@ -2391,9 +2339,7 @@
2391
  "model.layers.4.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
2392
  "model.layers.4.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
2393
  "model.layers.4.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
2394
- "model.layers.4.block_sparse_moe.gate.weight_packed": "model-00001-of-00005.safetensors",
2395
- "model.layers.4.block_sparse_moe.gate.weight_scale": "model-00001-of-00005.safetensors",
2396
- "model.layers.4.block_sparse_moe.gate.weight_shape": "model-00001-of-00005.safetensors",
2397
  "model.layers.4.input_layernorm.weight": "model-00001-of-00005.safetensors",
2398
  "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
2399
  "model.layers.4.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
@@ -2480,9 +2426,7 @@
2480
  "model.layers.5.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
2481
  "model.layers.5.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
2482
  "model.layers.5.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
2483
- "model.layers.5.block_sparse_moe.gate.weight_packed": "model-00001-of-00005.safetensors",
2484
- "model.layers.5.block_sparse_moe.gate.weight_scale": "model-00001-of-00005.safetensors",
2485
- "model.layers.5.block_sparse_moe.gate.weight_shape": "model-00001-of-00005.safetensors",
2486
  "model.layers.5.input_layernorm.weight": "model-00001-of-00005.safetensors",
2487
  "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
2488
  "model.layers.5.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
@@ -2569,9 +2513,7 @@
2569
  "model.layers.6.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
2570
  "model.layers.6.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
2571
  "model.layers.6.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
2572
- "model.layers.6.block_sparse_moe.gate.weight_packed": "model-00001-of-00005.safetensors",
2573
- "model.layers.6.block_sparse_moe.gate.weight_scale": "model-00001-of-00005.safetensors",
2574
- "model.layers.6.block_sparse_moe.gate.weight_shape": "model-00001-of-00005.safetensors",
2575
  "model.layers.6.input_layernorm.weight": "model-00002-of-00005.safetensors",
2576
  "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
2577
  "model.layers.6.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
@@ -2658,9 +2600,7 @@
2658
  "model.layers.7.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
2659
  "model.layers.7.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
2660
  "model.layers.7.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
2661
- "model.layers.7.block_sparse_moe.gate.weight_packed": "model-00002-of-00005.safetensors",
2662
- "model.layers.7.block_sparse_moe.gate.weight_scale": "model-00002-of-00005.safetensors",
2663
- "model.layers.7.block_sparse_moe.gate.weight_shape": "model-00002-of-00005.safetensors",
2664
  "model.layers.7.input_layernorm.weight": "model-00002-of-00005.safetensors",
2665
  "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
2666
  "model.layers.7.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
@@ -2747,9 +2687,7 @@
2747
  "model.layers.8.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
2748
  "model.layers.8.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
2749
  "model.layers.8.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
2750
- "model.layers.8.block_sparse_moe.gate.weight_packed": "model-00002-of-00005.safetensors",
2751
- "model.layers.8.block_sparse_moe.gate.weight_scale": "model-00002-of-00005.safetensors",
2752
- "model.layers.8.block_sparse_moe.gate.weight_shape": "model-00002-of-00005.safetensors",
2753
  "model.layers.8.input_layernorm.weight": "model-00002-of-00005.safetensors",
2754
  "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
2755
  "model.layers.8.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
@@ -2836,9 +2774,7 @@
2836
  "model.layers.9.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
2837
  "model.layers.9.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
2838
  "model.layers.9.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
2839
- "model.layers.9.block_sparse_moe.gate.weight_packed": "model-00002-of-00005.safetensors",
2840
- "model.layers.9.block_sparse_moe.gate.weight_scale": "model-00002-of-00005.safetensors",
2841
- "model.layers.9.block_sparse_moe.gate.weight_shape": "model-00002-of-00005.safetensors",
2842
  "model.layers.9.input_layernorm.weight": "model-00002-of-00005.safetensors",
2843
  "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
2844
  "model.layers.9.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
 
1
  {
2
  "metadata": {
3
+ "total_size": 24472213504
4
  },
5
  "weight_map": {
6
  "lm_head.weight": "model-00005-of-00005.safetensors",
 
77
  "model.layers.0.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
78
  "model.layers.0.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
79
  "model.layers.0.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
80
+ "model.layers.0.block_sparse_moe.gate.weight": "model-00001-of-00005.safetensors",
 
 
81
  "model.layers.0.input_layernorm.weight": "model-00001-of-00005.safetensors",
82
  "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
83
  "model.layers.0.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
 
164
  "model.layers.1.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
165
  "model.layers.1.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
166
  "model.layers.1.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
167
+ "model.layers.1.block_sparse_moe.gate.weight": "model-00001-of-00005.safetensors",
 
 
168
  "model.layers.1.input_layernorm.weight": "model-00001-of-00005.safetensors",
169
  "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
170
  "model.layers.1.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
 
251
  "model.layers.10.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
252
  "model.layers.10.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
253
  "model.layers.10.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
254
+ "model.layers.10.block_sparse_moe.gate.weight": "model-00002-of-00005.safetensors",
 
 
255
  "model.layers.10.input_layernorm.weight": "model-00002-of-00005.safetensors",
256
  "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
257
  "model.layers.10.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
 
338
  "model.layers.11.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
339
  "model.layers.11.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
340
  "model.layers.11.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
341
+ "model.layers.11.block_sparse_moe.gate.weight": "model-00002-of-00005.safetensors",
 
 
342
  "model.layers.11.input_layernorm.weight": "model-00002-of-00005.safetensors",
343
  "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
344
  "model.layers.11.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
 
425
  "model.layers.12.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
426
  "model.layers.12.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
427
  "model.layers.12.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
428
+ "model.layers.12.block_sparse_moe.gate.weight": "model-00002-of-00005.safetensors",
 
 
429
  "model.layers.12.input_layernorm.weight": "model-00003-of-00005.safetensors",
430
  "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
431
  "model.layers.12.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
 
512
  "model.layers.13.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
513
  "model.layers.13.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
514
  "model.layers.13.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
515
+ "model.layers.13.block_sparse_moe.gate.weight": "model-00003-of-00005.safetensors",
 
 
516
  "model.layers.13.input_layernorm.weight": "model-00003-of-00005.safetensors",
517
  "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
518
  "model.layers.13.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
 
599
  "model.layers.14.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
600
  "model.layers.14.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
601
  "model.layers.14.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
602
+ "model.layers.14.block_sparse_moe.gate.weight": "model-00003-of-00005.safetensors",
 
 
603
  "model.layers.14.input_layernorm.weight": "model-00003-of-00005.safetensors",
604
  "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
605
  "model.layers.14.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
 
686
  "model.layers.15.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
687
  "model.layers.15.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
688
  "model.layers.15.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
689
+ "model.layers.15.block_sparse_moe.gate.weight": "model-00003-of-00005.safetensors",
 
 
690
  "model.layers.15.input_layernorm.weight": "model-00003-of-00005.safetensors",
691
  "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
692
  "model.layers.15.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
 
773
  "model.layers.16.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
774
  "model.layers.16.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
775
  "model.layers.16.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
776
+ "model.layers.16.block_sparse_moe.gate.weight": "model-00003-of-00005.safetensors",
 
 
777
  "model.layers.16.input_layernorm.weight": "model-00003-of-00005.safetensors",
778
  "model.layers.16.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
779
  "model.layers.16.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
 
860
  "model.layers.17.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
861
  "model.layers.17.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
862
  "model.layers.17.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
863
+ "model.layers.17.block_sparse_moe.gate.weight": "model-00003-of-00005.safetensors",
 
 
864
  "model.layers.17.input_layernorm.weight": "model-00003-of-00005.safetensors",
865
  "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
866
  "model.layers.17.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
 
947
  "model.layers.18.block_sparse_moe.experts.7.w3.weight_packed": "model-00003-of-00005.safetensors",
948
  "model.layers.18.block_sparse_moe.experts.7.w3.weight_scale": "model-00003-of-00005.safetensors",
949
  "model.layers.18.block_sparse_moe.experts.7.w3.weight_shape": "model-00003-of-00005.safetensors",
950
+ "model.layers.18.block_sparse_moe.gate.weight": "model-00003-of-00005.safetensors",
 
 
951
  "model.layers.18.input_layernorm.weight": "model-00003-of-00005.safetensors",
952
  "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
953
  "model.layers.18.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
 
1034
  "model.layers.19.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1035
  "model.layers.19.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1036
  "model.layers.19.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1037
+ "model.layers.19.block_sparse_moe.gate.weight": "model-00003-of-00005.safetensors",
 
 
1038
  "model.layers.19.input_layernorm.weight": "model-00004-of-00005.safetensors",
1039
  "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1040
  "model.layers.19.self_attn.k_proj.weight_packed": "model-00003-of-00005.safetensors",
 
1121
  "model.layers.2.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
1122
  "model.layers.2.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
1123
  "model.layers.2.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
1124
+ "model.layers.2.block_sparse_moe.gate.weight": "model-00001-of-00005.safetensors",
 
 
1125
  "model.layers.2.input_layernorm.weight": "model-00001-of-00005.safetensors",
1126
  "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
1127
  "model.layers.2.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
 
1208
  "model.layers.20.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1209
  "model.layers.20.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1210
  "model.layers.20.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1211
+ "model.layers.20.block_sparse_moe.gate.weight": "model-00004-of-00005.safetensors",
 
 
1212
  "model.layers.20.input_layernorm.weight": "model-00004-of-00005.safetensors",
1213
  "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1214
  "model.layers.20.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
 
1295
  "model.layers.21.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1296
  "model.layers.21.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1297
  "model.layers.21.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1298
+ "model.layers.21.block_sparse_moe.gate.weight": "model-00004-of-00005.safetensors",
 
 
1299
  "model.layers.21.input_layernorm.weight": "model-00004-of-00005.safetensors",
1300
  "model.layers.21.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1301
  "model.layers.21.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
 
1382
  "model.layers.22.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1383
  "model.layers.22.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1384
  "model.layers.22.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1385
+ "model.layers.22.block_sparse_moe.gate.weight": "model-00004-of-00005.safetensors",
 
 
1386
  "model.layers.22.input_layernorm.weight": "model-00004-of-00005.safetensors",
1387
  "model.layers.22.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1388
  "model.layers.22.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
 
1469
  "model.layers.23.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1470
  "model.layers.23.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1471
  "model.layers.23.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1472
+ "model.layers.23.block_sparse_moe.gate.weight": "model-00004-of-00005.safetensors",
 
 
1473
  "model.layers.23.input_layernorm.weight": "model-00004-of-00005.safetensors",
1474
  "model.layers.23.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1475
  "model.layers.23.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
 
1556
  "model.layers.24.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1557
  "model.layers.24.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1558
  "model.layers.24.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1559
+ "model.layers.24.block_sparse_moe.gate.weight": "model-00004-of-00005.safetensors",
 
 
1560
  "model.layers.24.input_layernorm.weight": "model-00004-of-00005.safetensors",
1561
  "model.layers.24.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1562
  "model.layers.24.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
 
1643
  "model.layers.25.block_sparse_moe.experts.7.w3.weight_packed": "model-00004-of-00005.safetensors",
1644
  "model.layers.25.block_sparse_moe.experts.7.w3.weight_scale": "model-00004-of-00005.safetensors",
1645
  "model.layers.25.block_sparse_moe.experts.7.w3.weight_shape": "model-00004-of-00005.safetensors",
1646
+ "model.layers.25.block_sparse_moe.gate.weight": "model-00004-of-00005.safetensors",
 
 
1647
  "model.layers.25.input_layernorm.weight": "model-00004-of-00005.safetensors",
1648
  "model.layers.25.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
1649
  "model.layers.25.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
 
1730
  "model.layers.26.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
1731
  "model.layers.26.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
1732
  "model.layers.26.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
1733
+ "model.layers.26.block_sparse_moe.gate.weight": "model-00004-of-00005.safetensors",
 
 
1734
  "model.layers.26.input_layernorm.weight": "model-00005-of-00005.safetensors",
1735
  "model.layers.26.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
1736
  "model.layers.26.self_attn.k_proj.weight_packed": "model-00004-of-00005.safetensors",
 
1817
  "model.layers.27.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
1818
  "model.layers.27.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
1819
  "model.layers.27.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
1820
+ "model.layers.27.block_sparse_moe.gate.weight": "model-00005-of-00005.safetensors",
 
 
1821
  "model.layers.27.input_layernorm.weight": "model-00005-of-00005.safetensors",
1822
  "model.layers.27.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
1823
  "model.layers.27.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
 
1904
  "model.layers.28.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
1905
  "model.layers.28.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
1906
  "model.layers.28.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
1907
+ "model.layers.28.block_sparse_moe.gate.weight": "model-00005-of-00005.safetensors",
 
 
1908
  "model.layers.28.input_layernorm.weight": "model-00005-of-00005.safetensors",
1909
  "model.layers.28.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
1910
  "model.layers.28.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
 
1991
  "model.layers.29.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
1992
  "model.layers.29.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
1993
  "model.layers.29.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
1994
+ "model.layers.29.block_sparse_moe.gate.weight": "model-00005-of-00005.safetensors",
 
 
1995
  "model.layers.29.input_layernorm.weight": "model-00005-of-00005.safetensors",
1996
  "model.layers.29.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
1997
  "model.layers.29.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
 
2078
  "model.layers.3.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
2079
  "model.layers.3.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
2080
  "model.layers.3.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
2081
+ "model.layers.3.block_sparse_moe.gate.weight": "model-00001-of-00005.safetensors",
 
 
2082
  "model.layers.3.input_layernorm.weight": "model-00001-of-00005.safetensors",
2083
  "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
2084
  "model.layers.3.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
 
2165
  "model.layers.30.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
2166
  "model.layers.30.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
2167
  "model.layers.30.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
2168
+ "model.layers.30.block_sparse_moe.gate.weight": "model-00005-of-00005.safetensors",
 
 
2169
  "model.layers.30.input_layernorm.weight": "model-00005-of-00005.safetensors",
2170
  "model.layers.30.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
2171
  "model.layers.30.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
 
2252
  "model.layers.31.block_sparse_moe.experts.7.w3.weight_packed": "model-00005-of-00005.safetensors",
2253
  "model.layers.31.block_sparse_moe.experts.7.w3.weight_scale": "model-00005-of-00005.safetensors",
2254
  "model.layers.31.block_sparse_moe.experts.7.w3.weight_shape": "model-00005-of-00005.safetensors",
2255
+ "model.layers.31.block_sparse_moe.gate.weight": "model-00005-of-00005.safetensors",
 
 
2256
  "model.layers.31.input_layernorm.weight": "model-00005-of-00005.safetensors",
2257
  "model.layers.31.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
2258
  "model.layers.31.self_attn.k_proj.weight_packed": "model-00005-of-00005.safetensors",
 
2339
  "model.layers.4.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
2340
  "model.layers.4.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
2341
  "model.layers.4.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
2342
+ "model.layers.4.block_sparse_moe.gate.weight": "model-00001-of-00005.safetensors",
 
 
2343
  "model.layers.4.input_layernorm.weight": "model-00001-of-00005.safetensors",
2344
  "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
2345
  "model.layers.4.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
 
2426
  "model.layers.5.block_sparse_moe.experts.7.w3.weight_packed": "model-00001-of-00005.safetensors",
2427
  "model.layers.5.block_sparse_moe.experts.7.w3.weight_scale": "model-00001-of-00005.safetensors",
2428
  "model.layers.5.block_sparse_moe.experts.7.w3.weight_shape": "model-00001-of-00005.safetensors",
2429
+ "model.layers.5.block_sparse_moe.gate.weight": "model-00001-of-00005.safetensors",
 
 
2430
  "model.layers.5.input_layernorm.weight": "model-00001-of-00005.safetensors",
2431
  "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
2432
  "model.layers.5.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
 
2513
  "model.layers.6.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
2514
  "model.layers.6.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
2515
  "model.layers.6.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
2516
+ "model.layers.6.block_sparse_moe.gate.weight": "model-00001-of-00005.safetensors",
 
 
2517
  "model.layers.6.input_layernorm.weight": "model-00002-of-00005.safetensors",
2518
  "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
2519
  "model.layers.6.self_attn.k_proj.weight_packed": "model-00001-of-00005.safetensors",
 
2600
  "model.layers.7.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
2601
  "model.layers.7.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
2602
  "model.layers.7.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
2603
+ "model.layers.7.block_sparse_moe.gate.weight": "model-00002-of-00005.safetensors",
 
 
2604
  "model.layers.7.input_layernorm.weight": "model-00002-of-00005.safetensors",
2605
  "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
2606
  "model.layers.7.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
 
2687
  "model.layers.8.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
2688
  "model.layers.8.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
2689
  "model.layers.8.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
2690
+ "model.layers.8.block_sparse_moe.gate.weight": "model-00002-of-00005.safetensors",
 
 
2691
  "model.layers.8.input_layernorm.weight": "model-00002-of-00005.safetensors",
2692
  "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
2693
  "model.layers.8.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
 
2774
  "model.layers.9.block_sparse_moe.experts.7.w3.weight_packed": "model-00002-of-00005.safetensors",
2775
  "model.layers.9.block_sparse_moe.experts.7.w3.weight_scale": "model-00002-of-00005.safetensors",
2776
  "model.layers.9.block_sparse_moe.experts.7.w3.weight_shape": "model-00002-of-00005.safetensors",
2777
+ "model.layers.9.block_sparse_moe.gate.weight": "model-00002-of-00005.safetensors",
 
 
2778
  "model.layers.9.input_layernorm.weight": "model-00002-of-00005.safetensors",
2779
  "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
2780
  "model.layers.9.self_attn.k_proj.weight_packed": "model-00002-of-00005.safetensors",
recipe.yaml CHANGED
@@ -3,5 +3,5 @@ DEFAULT_stage:
3
  GPTQModifier:
4
  sequential_update: true
5
  targets: Linear
6
- ignore: [lm_head]
7
  scheme: W4A16
 
3
  GPTQModifier:
4
  sequential_update: true
5
  targets: Linear
6
+ ignore: [lm_head, 're:.*block_sparse_moe\.gate']
7
  scheme: W4A16