BlackSamorez commited on
Commit
5133b63
1 Parent(s): ea1c68f

correct dtype + new dispatch + tokenizer

Browse files
config.json CHANGED
@@ -1,19 +1,9 @@
1
  {
2
- "_name_or_path": "aqml_mixtral_instruct",
3
- "aqlm": {
4
- "in_group_size": 8,
5
- "nbits_per_codebook": 16,
6
- "num_codebooks": 1,
7
- "out_group_size": 1
8
- },
9
  "architectures": [
10
  "MixtralForCausalLM"
11
  ],
12
  "attention_dropout": 0.0,
13
- "auto_map": {
14
- "AutoConfig": "configuration_mixtral_aqlm.MixtralConfig",
15
- "AutoModelForCausalLM": "modeling_mixtral_aqlm.MixtralForCausalLM"
16
- },
17
  "bos_token_id": 1,
18
  "eos_token_id": 2,
19
  "hidden_act": "silu",
@@ -21,20 +11,128 @@
21
  "initializer_range": 0.02,
22
  "intermediate_size": 14336,
23
  "max_position_embeddings": 32768,
24
- "model_type": "mixtral_aqlm",
25
  "num_attention_heads": 32,
26
  "num_experts_per_tok": 2,
27
  "num_hidden_layers": 32,
28
  "num_key_value_heads": 8,
29
  "num_local_experts": 8,
30
  "output_router_logits": false,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
31
  "rms_norm_eps": 1e-05,
32
  "rope_theta": 1000000.0,
33
  "router_aux_loss_coef": 0.02,
34
  "sliding_window": null,
35
  "tie_word_embeddings": false,
36
- "torch_dtype": "float32",
37
- "transformers_version": "4.37.2",
38
  "use_cache": true,
39
  "vocab_size": 32000
40
  }
 
1
  {
2
+ "_name_or_path": "BlackSamorez/Mixtral-8x7B-Instruct-v0_1-AQLM-2Bit-1x16-hf",
 
 
 
 
 
 
3
  "architectures": [
4
  "MixtralForCausalLM"
5
  ],
6
  "attention_dropout": 0.0,
 
 
 
 
7
  "bos_token_id": 1,
8
  "eos_token_id": 2,
9
  "hidden_act": "silu",
 
11
  "initializer_range": 0.02,
12
  "intermediate_size": 14336,
13
  "max_position_embeddings": 32768,
14
+ "model_type": "mixtral",
15
  "num_attention_heads": 32,
16
  "num_experts_per_tok": 2,
17
  "num_hidden_layers": 32,
18
  "num_key_value_heads": 8,
19
  "num_local_experts": 8,
20
  "output_router_logits": false,
21
+ "quantization_config": {
22
+ "in_group_size": 8,
23
+ "linear_weights_not_to_quantize": [
24
+ "model.layers.0.block_sparse_moe.gate.weight",
25
+ "model.layers.0.input_layernorm.weight",
26
+ "model.layers.0.post_attention_layernorm.weight",
27
+ "model.layers.1.block_sparse_moe.gate.weight",
28
+ "model.layers.1.input_layernorm.weight",
29
+ "model.layers.1.post_attention_layernorm.weight",
30
+ "model.layers.2.block_sparse_moe.gate.weight",
31
+ "model.layers.2.input_layernorm.weight",
32
+ "model.layers.2.post_attention_layernorm.weight",
33
+ "model.layers.3.block_sparse_moe.gate.weight",
34
+ "model.layers.3.input_layernorm.weight",
35
+ "model.layers.3.post_attention_layernorm.weight",
36
+ "model.layers.4.block_sparse_moe.gate.weight",
37
+ "model.layers.4.input_layernorm.weight",
38
+ "model.layers.4.post_attention_layernorm.weight",
39
+ "model.layers.5.block_sparse_moe.gate.weight",
40
+ "model.layers.5.input_layernorm.weight",
41
+ "model.layers.5.post_attention_layernorm.weight",
42
+ "model.layers.6.block_sparse_moe.gate.weight",
43
+ "model.layers.6.input_layernorm.weight",
44
+ "model.layers.6.post_attention_layernorm.weight",
45
+ "model.layers.7.block_sparse_moe.gate.weight",
46
+ "model.layers.7.input_layernorm.weight",
47
+ "model.layers.7.post_attention_layernorm.weight",
48
+ "model.layers.8.block_sparse_moe.gate.weight",
49
+ "model.layers.8.input_layernorm.weight",
50
+ "model.layers.8.post_attention_layernorm.weight",
51
+ "model.layers.9.block_sparse_moe.gate.weight",
52
+ "model.layers.9.input_layernorm.weight",
53
+ "model.layers.9.post_attention_layernorm.weight",
54
+ "model.layers.10.block_sparse_moe.gate.weight",
55
+ "model.layers.10.input_layernorm.weight",
56
+ "model.layers.10.post_attention_layernorm.weight",
57
+ "model.layers.11.block_sparse_moe.gate.weight",
58
+ "model.layers.11.input_layernorm.weight",
59
+ "model.layers.11.post_attention_layernorm.weight",
60
+ "model.layers.12.block_sparse_moe.gate.weight",
61
+ "model.layers.12.input_layernorm.weight",
62
+ "model.layers.12.post_attention_layernorm.weight",
63
+ "model.layers.13.block_sparse_moe.gate.weight",
64
+ "model.layers.13.input_layernorm.weight",
65
+ "model.layers.13.post_attention_layernorm.weight",
66
+ "model.layers.14.block_sparse_moe.gate.weight",
67
+ "model.layers.14.input_layernorm.weight",
68
+ "model.layers.14.post_attention_layernorm.weight",
69
+ "model.layers.15.block_sparse_moe.gate.weight",
70
+ "model.layers.15.input_layernorm.weight",
71
+ "model.layers.15.post_attention_layernorm.weight",
72
+ "model.layers.16.block_sparse_moe.gate.weight",
73
+ "model.layers.16.input_layernorm.weight",
74
+ "model.layers.16.post_attention_layernorm.weight",
75
+ "model.layers.17.block_sparse_moe.gate.weight",
76
+ "model.layers.17.input_layernorm.weight",
77
+ "model.layers.17.post_attention_layernorm.weight",
78
+ "model.layers.18.block_sparse_moe.gate.weight",
79
+ "model.layers.18.input_layernorm.weight",
80
+ "model.layers.18.post_attention_layernorm.weight",
81
+ "model.layers.19.block_sparse_moe.gate.weight",
82
+ "model.layers.19.input_layernorm.weight",
83
+ "model.layers.19.post_attention_layernorm.weight",
84
+ "model.layers.20.block_sparse_moe.gate.weight",
85
+ "model.layers.20.input_layernorm.weight",
86
+ "model.layers.20.post_attention_layernorm.weight",
87
+ "model.layers.21.block_sparse_moe.gate.weight",
88
+ "model.layers.21.input_layernorm.weight",
89
+ "model.layers.21.post_attention_layernorm.weight",
90
+ "model.layers.22.block_sparse_moe.gate.weight",
91
+ "model.layers.22.input_layernorm.weight",
92
+ "model.layers.22.post_attention_layernorm.weight",
93
+ "model.layers.23.block_sparse_moe.gate.weight",
94
+ "model.layers.23.input_layernorm.weight",
95
+ "model.layers.23.post_attention_layernorm.weight",
96
+ "model.layers.24.block_sparse_moe.gate.weight",
97
+ "model.layers.24.input_layernorm.weight",
98
+ "model.layers.24.post_attention_layernorm.weight",
99
+ "model.layers.25.block_sparse_moe.gate.weight",
100
+ "model.layers.25.input_layernorm.weight",
101
+ "model.layers.25.post_attention_layernorm.weight",
102
+ "model.layers.26.block_sparse_moe.gate.weight",
103
+ "model.layers.26.input_layernorm.weight",
104
+ "model.layers.26.post_attention_layernorm.weight",
105
+ "model.layers.27.block_sparse_moe.gate.weight",
106
+ "model.layers.27.input_layernorm.weight",
107
+ "model.layers.27.post_attention_layernorm.weight",
108
+ "model.layers.28.block_sparse_moe.gate.weight",
109
+ "model.layers.28.input_layernorm.weight",
110
+ "model.layers.28.post_attention_layernorm.weight",
111
+ "model.layers.29.block_sparse_moe.gate.weight",
112
+ "model.layers.29.input_layernorm.weight",
113
+ "model.layers.29.post_attention_layernorm.weight",
114
+ "model.layers.30.block_sparse_moe.gate.weight",
115
+ "model.layers.30.input_layernorm.weight",
116
+ "model.layers.30.post_attention_layernorm.weight",
117
+ "model.layers.31.block_sparse_moe.gate.weight",
118
+ "model.layers.31.input_layernorm.weight",
119
+ "model.layers.31.post_attention_layernorm.weight",
120
+ "model.embed_tokens.weight",
121
+ "model.norm.weight",
122
+ "lm_head.weight"
123
+ ],
124
+ "nbits_per_codebook": 16,
125
+ "num_codebooks": 1,
126
+ "out_group_size": 1,
127
+ "quant_method": "aqlm"
128
+ },
129
  "rms_norm_eps": 1e-05,
130
  "rope_theta": 1000000.0,
131
  "router_aux_loss_coef": 0.02,
132
  "sliding_window": null,
133
  "tie_word_embeddings": false,
134
+ "torch_dtype": "float16",
135
+ "transformers_version": "4.38.0",
136
  "use_cache": true,
137
  "vocab_size": 32000
138
  }
generation_config.json CHANGED
@@ -2,5 +2,5 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
- "transformers_version": "4.37.2"
6
  }
 
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
+ "transformers_version": "4.38.0"
6
  }
model-00001-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cf8bc80ab1ec1cbc52b400900ad7ecd5aaef2bcc7d88db740fa3760cd9822b72
3
- size 4992044680
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0852c33bc7857d0f2c353ca3b2c1055d8c682313042592c87cde4b770815325f
3
+ size 4991623320
model-00002-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a66a5755c322366e9da629b689ce6f68986a85cfdb8900e1d7e630705a3104b0
3
- size 4989371552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:798e45ce69ae4a62c4979d59ce6ac68b1a361ace6155864cbecb14e499b90cd9
3
+ size 4997179136
model-00003-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:abf1ff3b6c4e360f6668ffd8cc09c9ac2e521a3ad62635d7bc2043779c583c7b
3
- size 4596524152
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00801a120346eb99524095af4bde0d2885881feaa29d8b6bb15005b17e044f85
3
+ size 3105262720
model.safetensors.index.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "metadata": {
3
- "total_size": 14577582080
4
  },
5
  "weight_map": {
6
  "lm_head.weight": "model-00003-of-00003.safetensors",
@@ -219,41 +219,41 @@
219
  "model.layers.10.block_sparse_moe.experts.4.w1.codes": "model-00001-of-00003.safetensors",
220
  "model.layers.10.block_sparse_moe.experts.4.w1.scales": "model-00001-of-00003.safetensors",
221
  "model.layers.10.block_sparse_moe.experts.4.w2.codebooks": "model-00001-of-00003.safetensors",
222
- "model.layers.10.block_sparse_moe.experts.4.w2.codes": "model-00002-of-00003.safetensors",
223
- "model.layers.10.block_sparse_moe.experts.4.w2.scales": "model-00002-of-00003.safetensors",
224
- "model.layers.10.block_sparse_moe.experts.4.w3.codebooks": "model-00002-of-00003.safetensors",
225
- "model.layers.10.block_sparse_moe.experts.4.w3.codes": "model-00002-of-00003.safetensors",
226
- "model.layers.10.block_sparse_moe.experts.4.w3.scales": "model-00002-of-00003.safetensors",
227
- "model.layers.10.block_sparse_moe.experts.5.w1.codebooks": "model-00002-of-00003.safetensors",
228
- "model.layers.10.block_sparse_moe.experts.5.w1.codes": "model-00002-of-00003.safetensors",
229
- "model.layers.10.block_sparse_moe.experts.5.w1.scales": "model-00002-of-00003.safetensors",
230
- "model.layers.10.block_sparse_moe.experts.5.w2.codebooks": "model-00002-of-00003.safetensors",
231
- "model.layers.10.block_sparse_moe.experts.5.w2.codes": "model-00002-of-00003.safetensors",
232
- "model.layers.10.block_sparse_moe.experts.5.w2.scales": "model-00002-of-00003.safetensors",
233
- "model.layers.10.block_sparse_moe.experts.5.w3.codebooks": "model-00002-of-00003.safetensors",
234
- "model.layers.10.block_sparse_moe.experts.5.w3.codes": "model-00002-of-00003.safetensors",
235
- "model.layers.10.block_sparse_moe.experts.5.w3.scales": "model-00002-of-00003.safetensors",
236
- "model.layers.10.block_sparse_moe.experts.6.w1.codebooks": "model-00002-of-00003.safetensors",
237
- "model.layers.10.block_sparse_moe.experts.6.w1.codes": "model-00002-of-00003.safetensors",
238
- "model.layers.10.block_sparse_moe.experts.6.w1.scales": "model-00002-of-00003.safetensors",
239
- "model.layers.10.block_sparse_moe.experts.6.w2.codebooks": "model-00002-of-00003.safetensors",
240
- "model.layers.10.block_sparse_moe.experts.6.w2.codes": "model-00002-of-00003.safetensors",
241
- "model.layers.10.block_sparse_moe.experts.6.w2.scales": "model-00002-of-00003.safetensors",
242
- "model.layers.10.block_sparse_moe.experts.6.w3.codebooks": "model-00002-of-00003.safetensors",
243
- "model.layers.10.block_sparse_moe.experts.6.w3.codes": "model-00002-of-00003.safetensors",
244
- "model.layers.10.block_sparse_moe.experts.6.w3.scales": "model-00002-of-00003.safetensors",
245
- "model.layers.10.block_sparse_moe.experts.7.w1.codebooks": "model-00002-of-00003.safetensors",
246
- "model.layers.10.block_sparse_moe.experts.7.w1.codes": "model-00002-of-00003.safetensors",
247
- "model.layers.10.block_sparse_moe.experts.7.w1.scales": "model-00002-of-00003.safetensors",
248
- "model.layers.10.block_sparse_moe.experts.7.w2.codebooks": "model-00002-of-00003.safetensors",
249
- "model.layers.10.block_sparse_moe.experts.7.w2.codes": "model-00002-of-00003.safetensors",
250
- "model.layers.10.block_sparse_moe.experts.7.w2.scales": "model-00002-of-00003.safetensors",
251
- "model.layers.10.block_sparse_moe.experts.7.w3.codebooks": "model-00002-of-00003.safetensors",
252
- "model.layers.10.block_sparse_moe.experts.7.w3.codes": "model-00002-of-00003.safetensors",
253
- "model.layers.10.block_sparse_moe.experts.7.w3.scales": "model-00002-of-00003.safetensors",
254
  "model.layers.10.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
255
- "model.layers.10.input_layernorm.weight": "model-00002-of-00003.safetensors",
256
- "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
257
  "model.layers.10.self_attn.k_proj.codebooks": "model-00001-of-00003.safetensors",
258
  "model.layers.10.self_attn.k_proj.codes": "model-00001-of-00003.safetensors",
259
  "model.layers.10.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
@@ -266,94 +266,94 @@
266
  "model.layers.10.self_attn.v_proj.codebooks": "model-00001-of-00003.safetensors",
267
  "model.layers.10.self_attn.v_proj.codes": "model-00001-of-00003.safetensors",
268
  "model.layers.10.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
269
- "model.layers.11.block_sparse_moe.experts.0.w1.codebooks": "model-00002-of-00003.safetensors",
270
- "model.layers.11.block_sparse_moe.experts.0.w1.codes": "model-00002-of-00003.safetensors",
271
- "model.layers.11.block_sparse_moe.experts.0.w1.scales": "model-00002-of-00003.safetensors",
272
- "model.layers.11.block_sparse_moe.experts.0.w2.codebooks": "model-00002-of-00003.safetensors",
273
- "model.layers.11.block_sparse_moe.experts.0.w2.codes": "model-00002-of-00003.safetensors",
274
- "model.layers.11.block_sparse_moe.experts.0.w2.scales": "model-00002-of-00003.safetensors",
275
- "model.layers.11.block_sparse_moe.experts.0.w3.codebooks": "model-00002-of-00003.safetensors",
276
- "model.layers.11.block_sparse_moe.experts.0.w3.codes": "model-00002-of-00003.safetensors",
277
- "model.layers.11.block_sparse_moe.experts.0.w3.scales": "model-00002-of-00003.safetensors",
278
- "model.layers.11.block_sparse_moe.experts.1.w1.codebooks": "model-00002-of-00003.safetensors",
279
- "model.layers.11.block_sparse_moe.experts.1.w1.codes": "model-00002-of-00003.safetensors",
280
- "model.layers.11.block_sparse_moe.experts.1.w1.scales": "model-00002-of-00003.safetensors",
281
- "model.layers.11.block_sparse_moe.experts.1.w2.codebooks": "model-00002-of-00003.safetensors",
282
- "model.layers.11.block_sparse_moe.experts.1.w2.codes": "model-00002-of-00003.safetensors",
283
- "model.layers.11.block_sparse_moe.experts.1.w2.scales": "model-00002-of-00003.safetensors",
284
- "model.layers.11.block_sparse_moe.experts.1.w3.codebooks": "model-00002-of-00003.safetensors",
285
- "model.layers.11.block_sparse_moe.experts.1.w3.codes": "model-00002-of-00003.safetensors",
286
- "model.layers.11.block_sparse_moe.experts.1.w3.scales": "model-00002-of-00003.safetensors",
287
- "model.layers.11.block_sparse_moe.experts.2.w1.codebooks": "model-00002-of-00003.safetensors",
288
- "model.layers.11.block_sparse_moe.experts.2.w1.codes": "model-00002-of-00003.safetensors",
289
- "model.layers.11.block_sparse_moe.experts.2.w1.scales": "model-00002-of-00003.safetensors",
290
- "model.layers.11.block_sparse_moe.experts.2.w2.codebooks": "model-00002-of-00003.safetensors",
291
- "model.layers.11.block_sparse_moe.experts.2.w2.codes": "model-00002-of-00003.safetensors",
292
- "model.layers.11.block_sparse_moe.experts.2.w2.scales": "model-00002-of-00003.safetensors",
293
- "model.layers.11.block_sparse_moe.experts.2.w3.codebooks": "model-00002-of-00003.safetensors",
294
- "model.layers.11.block_sparse_moe.experts.2.w3.codes": "model-00002-of-00003.safetensors",
295
- "model.layers.11.block_sparse_moe.experts.2.w3.scales": "model-00002-of-00003.safetensors",
296
- "model.layers.11.block_sparse_moe.experts.3.w1.codebooks": "model-00002-of-00003.safetensors",
297
- "model.layers.11.block_sparse_moe.experts.3.w1.codes": "model-00002-of-00003.safetensors",
298
- "model.layers.11.block_sparse_moe.experts.3.w1.scales": "model-00002-of-00003.safetensors",
299
- "model.layers.11.block_sparse_moe.experts.3.w2.codebooks": "model-00002-of-00003.safetensors",
300
- "model.layers.11.block_sparse_moe.experts.3.w2.codes": "model-00002-of-00003.safetensors",
301
- "model.layers.11.block_sparse_moe.experts.3.w2.scales": "model-00002-of-00003.safetensors",
302
- "model.layers.11.block_sparse_moe.experts.3.w3.codebooks": "model-00002-of-00003.safetensors",
303
- "model.layers.11.block_sparse_moe.experts.3.w3.codes": "model-00002-of-00003.safetensors",
304
- "model.layers.11.block_sparse_moe.experts.3.w3.scales": "model-00002-of-00003.safetensors",
305
- "model.layers.11.block_sparse_moe.experts.4.w1.codebooks": "model-00002-of-00003.safetensors",
306
- "model.layers.11.block_sparse_moe.experts.4.w1.codes": "model-00002-of-00003.safetensors",
307
- "model.layers.11.block_sparse_moe.experts.4.w1.scales": "model-00002-of-00003.safetensors",
308
- "model.layers.11.block_sparse_moe.experts.4.w2.codebooks": "model-00002-of-00003.safetensors",
309
- "model.layers.11.block_sparse_moe.experts.4.w2.codes": "model-00002-of-00003.safetensors",
310
- "model.layers.11.block_sparse_moe.experts.4.w2.scales": "model-00002-of-00003.safetensors",
311
- "model.layers.11.block_sparse_moe.experts.4.w3.codebooks": "model-00002-of-00003.safetensors",
312
- "model.layers.11.block_sparse_moe.experts.4.w3.codes": "model-00002-of-00003.safetensors",
313
- "model.layers.11.block_sparse_moe.experts.4.w3.scales": "model-00002-of-00003.safetensors",
314
- "model.layers.11.block_sparse_moe.experts.5.w1.codebooks": "model-00002-of-00003.safetensors",
315
- "model.layers.11.block_sparse_moe.experts.5.w1.codes": "model-00002-of-00003.safetensors",
316
- "model.layers.11.block_sparse_moe.experts.5.w1.scales": "model-00002-of-00003.safetensors",
317
- "model.layers.11.block_sparse_moe.experts.5.w2.codebooks": "model-00002-of-00003.safetensors",
318
- "model.layers.11.block_sparse_moe.experts.5.w2.codes": "model-00002-of-00003.safetensors",
319
- "model.layers.11.block_sparse_moe.experts.5.w2.scales": "model-00002-of-00003.safetensors",
320
- "model.layers.11.block_sparse_moe.experts.5.w3.codebooks": "model-00002-of-00003.safetensors",
321
- "model.layers.11.block_sparse_moe.experts.5.w3.codes": "model-00002-of-00003.safetensors",
322
- "model.layers.11.block_sparse_moe.experts.5.w3.scales": "model-00002-of-00003.safetensors",
323
- "model.layers.11.block_sparse_moe.experts.6.w1.codebooks": "model-00002-of-00003.safetensors",
324
- "model.layers.11.block_sparse_moe.experts.6.w1.codes": "model-00002-of-00003.safetensors",
325
- "model.layers.11.block_sparse_moe.experts.6.w1.scales": "model-00002-of-00003.safetensors",
326
- "model.layers.11.block_sparse_moe.experts.6.w2.codebooks": "model-00002-of-00003.safetensors",
327
- "model.layers.11.block_sparse_moe.experts.6.w2.codes": "model-00002-of-00003.safetensors",
328
- "model.layers.11.block_sparse_moe.experts.6.w2.scales": "model-00002-of-00003.safetensors",
329
- "model.layers.11.block_sparse_moe.experts.6.w3.codebooks": "model-00002-of-00003.safetensors",
330
- "model.layers.11.block_sparse_moe.experts.6.w3.codes": "model-00002-of-00003.safetensors",
331
- "model.layers.11.block_sparse_moe.experts.6.w3.scales": "model-00002-of-00003.safetensors",
332
- "model.layers.11.block_sparse_moe.experts.7.w1.codebooks": "model-00002-of-00003.safetensors",
333
- "model.layers.11.block_sparse_moe.experts.7.w1.codes": "model-00002-of-00003.safetensors",
334
- "model.layers.11.block_sparse_moe.experts.7.w1.scales": "model-00002-of-00003.safetensors",
335
- "model.layers.11.block_sparse_moe.experts.7.w2.codebooks": "model-00002-of-00003.safetensors",
336
- "model.layers.11.block_sparse_moe.experts.7.w2.codes": "model-00002-of-00003.safetensors",
337
- "model.layers.11.block_sparse_moe.experts.7.w2.scales": "model-00002-of-00003.safetensors",
338
- "model.layers.11.block_sparse_moe.experts.7.w3.codebooks": "model-00002-of-00003.safetensors",
339
- "model.layers.11.block_sparse_moe.experts.7.w3.codes": "model-00002-of-00003.safetensors",
340
- "model.layers.11.block_sparse_moe.experts.7.w3.scales": "model-00002-of-00003.safetensors",
341
- "model.layers.11.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
342
- "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
343
- "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
344
- "model.layers.11.self_attn.k_proj.codebooks": "model-00002-of-00003.safetensors",
345
- "model.layers.11.self_attn.k_proj.codes": "model-00002-of-00003.safetensors",
346
- "model.layers.11.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
347
- "model.layers.11.self_attn.o_proj.codebooks": "model-00002-of-00003.safetensors",
348
- "model.layers.11.self_attn.o_proj.codes": "model-00002-of-00003.safetensors",
349
- "model.layers.11.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
350
- "model.layers.11.self_attn.q_proj.codebooks": "model-00002-of-00003.safetensors",
351
- "model.layers.11.self_attn.q_proj.codes": "model-00002-of-00003.safetensors",
352
- "model.layers.11.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
353
- "model.layers.11.self_attn.v_proj.codebooks": "model-00002-of-00003.safetensors",
354
- "model.layers.11.self_attn.v_proj.codes": "model-00002-of-00003.safetensors",
355
- "model.layers.11.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
356
- "model.layers.12.block_sparse_moe.experts.0.w1.codebooks": "model-00002-of-00003.safetensors",
357
  "model.layers.12.block_sparse_moe.experts.0.w1.codes": "model-00002-of-00003.safetensors",
358
  "model.layers.12.block_sparse_moe.experts.0.w1.scales": "model-00002-of-00003.safetensors",
359
  "model.layers.12.block_sparse_moe.experts.0.w2.codebooks": "model-00002-of-00003.safetensors",
@@ -425,21 +425,21 @@
425
  "model.layers.12.block_sparse_moe.experts.7.w3.codebooks": "model-00002-of-00003.safetensors",
426
  "model.layers.12.block_sparse_moe.experts.7.w3.codes": "model-00002-of-00003.safetensors",
427
  "model.layers.12.block_sparse_moe.experts.7.w3.scales": "model-00002-of-00003.safetensors",
428
- "model.layers.12.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
429
  "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
430
  "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
431
- "model.layers.12.self_attn.k_proj.codebooks": "model-00002-of-00003.safetensors",
432
- "model.layers.12.self_attn.k_proj.codes": "model-00002-of-00003.safetensors",
433
- "model.layers.12.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
434
- "model.layers.12.self_attn.o_proj.codebooks": "model-00002-of-00003.safetensors",
435
- "model.layers.12.self_attn.o_proj.codes": "model-00002-of-00003.safetensors",
436
- "model.layers.12.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
437
- "model.layers.12.self_attn.q_proj.codebooks": "model-00002-of-00003.safetensors",
438
- "model.layers.12.self_attn.q_proj.codes": "model-00002-of-00003.safetensors",
439
- "model.layers.12.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
440
- "model.layers.12.self_attn.v_proj.codebooks": "model-00002-of-00003.safetensors",
441
- "model.layers.12.self_attn.v_proj.codes": "model-00002-of-00003.safetensors",
442
- "model.layers.12.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
443
  "model.layers.13.block_sparse_moe.experts.0.w1.codebooks": "model-00002-of-00003.safetensors",
444
  "model.layers.13.block_sparse_moe.experts.0.w1.codes": "model-00002-of-00003.safetensors",
445
  "model.layers.13.block_sparse_moe.experts.0.w1.scales": "model-00002-of-00003.safetensors",
@@ -1335,56 +1335,56 @@
1335
  "model.layers.22.block_sparse_moe.experts.2.w2.codes": "model-00002-of-00003.safetensors",
1336
  "model.layers.22.block_sparse_moe.experts.2.w2.scales": "model-00002-of-00003.safetensors",
1337
  "model.layers.22.block_sparse_moe.experts.2.w3.codebooks": "model-00002-of-00003.safetensors",
1338
- "model.layers.22.block_sparse_moe.experts.2.w3.codes": "model-00003-of-00003.safetensors",
1339
- "model.layers.22.block_sparse_moe.experts.2.w3.scales": "model-00003-of-00003.safetensors",
1340
- "model.layers.22.block_sparse_moe.experts.3.w1.codebooks": "model-00003-of-00003.safetensors",
1341
- "model.layers.22.block_sparse_moe.experts.3.w1.codes": "model-00003-of-00003.safetensors",
1342
- "model.layers.22.block_sparse_moe.experts.3.w1.scales": "model-00003-of-00003.safetensors",
1343
- "model.layers.22.block_sparse_moe.experts.3.w2.codebooks": "model-00003-of-00003.safetensors",
1344
- "model.layers.22.block_sparse_moe.experts.3.w2.codes": "model-00003-of-00003.safetensors",
1345
- "model.layers.22.block_sparse_moe.experts.3.w2.scales": "model-00003-of-00003.safetensors",
1346
- "model.layers.22.block_sparse_moe.experts.3.w3.codebooks": "model-00003-of-00003.safetensors",
1347
- "model.layers.22.block_sparse_moe.experts.3.w3.codes": "model-00003-of-00003.safetensors",
1348
- "model.layers.22.block_sparse_moe.experts.3.w3.scales": "model-00003-of-00003.safetensors",
1349
- "model.layers.22.block_sparse_moe.experts.4.w1.codebooks": "model-00003-of-00003.safetensors",
1350
- "model.layers.22.block_sparse_moe.experts.4.w1.codes": "model-00003-of-00003.safetensors",
1351
- "model.layers.22.block_sparse_moe.experts.4.w1.scales": "model-00003-of-00003.safetensors",
1352
- "model.layers.22.block_sparse_moe.experts.4.w2.codebooks": "model-00003-of-00003.safetensors",
1353
- "model.layers.22.block_sparse_moe.experts.4.w2.codes": "model-00003-of-00003.safetensors",
1354
- "model.layers.22.block_sparse_moe.experts.4.w2.scales": "model-00003-of-00003.safetensors",
1355
- "model.layers.22.block_sparse_moe.experts.4.w3.codebooks": "model-00003-of-00003.safetensors",
1356
- "model.layers.22.block_sparse_moe.experts.4.w3.codes": "model-00003-of-00003.safetensors",
1357
- "model.layers.22.block_sparse_moe.experts.4.w3.scales": "model-00003-of-00003.safetensors",
1358
- "model.layers.22.block_sparse_moe.experts.5.w1.codebooks": "model-00003-of-00003.safetensors",
1359
- "model.layers.22.block_sparse_moe.experts.5.w1.codes": "model-00003-of-00003.safetensors",
1360
- "model.layers.22.block_sparse_moe.experts.5.w1.scales": "model-00003-of-00003.safetensors",
1361
- "model.layers.22.block_sparse_moe.experts.5.w2.codebooks": "model-00003-of-00003.safetensors",
1362
- "model.layers.22.block_sparse_moe.experts.5.w2.codes": "model-00003-of-00003.safetensors",
1363
- "model.layers.22.block_sparse_moe.experts.5.w2.scales": "model-00003-of-00003.safetensors",
1364
- "model.layers.22.block_sparse_moe.experts.5.w3.codebooks": "model-00003-of-00003.safetensors",
1365
- "model.layers.22.block_sparse_moe.experts.5.w3.codes": "model-00003-of-00003.safetensors",
1366
- "model.layers.22.block_sparse_moe.experts.5.w3.scales": "model-00003-of-00003.safetensors",
1367
- "model.layers.22.block_sparse_moe.experts.6.w1.codebooks": "model-00003-of-00003.safetensors",
1368
- "model.layers.22.block_sparse_moe.experts.6.w1.codes": "model-00003-of-00003.safetensors",
1369
- "model.layers.22.block_sparse_moe.experts.6.w1.scales": "model-00003-of-00003.safetensors",
1370
- "model.layers.22.block_sparse_moe.experts.6.w2.codebooks": "model-00003-of-00003.safetensors",
1371
- "model.layers.22.block_sparse_moe.experts.6.w2.codes": "model-00003-of-00003.safetensors",
1372
- "model.layers.22.block_sparse_moe.experts.6.w2.scales": "model-00003-of-00003.safetensors",
1373
- "model.layers.22.block_sparse_moe.experts.6.w3.codebooks": "model-00003-of-00003.safetensors",
1374
- "model.layers.22.block_sparse_moe.experts.6.w3.codes": "model-00003-of-00003.safetensors",
1375
- "model.layers.22.block_sparse_moe.experts.6.w3.scales": "model-00003-of-00003.safetensors",
1376
- "model.layers.22.block_sparse_moe.experts.7.w1.codebooks": "model-00003-of-00003.safetensors",
1377
- "model.layers.22.block_sparse_moe.experts.7.w1.codes": "model-00003-of-00003.safetensors",
1378
- "model.layers.22.block_sparse_moe.experts.7.w1.scales": "model-00003-of-00003.safetensors",
1379
- "model.layers.22.block_sparse_moe.experts.7.w2.codebooks": "model-00003-of-00003.safetensors",
1380
- "model.layers.22.block_sparse_moe.experts.7.w2.codes": "model-00003-of-00003.safetensors",
1381
- "model.layers.22.block_sparse_moe.experts.7.w2.scales": "model-00003-of-00003.safetensors",
1382
- "model.layers.22.block_sparse_moe.experts.7.w3.codebooks": "model-00003-of-00003.safetensors",
1383
- "model.layers.22.block_sparse_moe.experts.7.w3.codes": "model-00003-of-00003.safetensors",
1384
- "model.layers.22.block_sparse_moe.experts.7.w3.scales": "model-00003-of-00003.safetensors",
1385
  "model.layers.22.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
1386
- "model.layers.22.input_layernorm.weight": "model-00003-of-00003.safetensors",
1387
- "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
1388
  "model.layers.22.self_attn.k_proj.codebooks": "model-00002-of-00003.safetensors",
1389
  "model.layers.22.self_attn.k_proj.codes": "model-00002-of-00003.safetensors",
1390
  "model.layers.22.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
@@ -1397,148 +1397,148 @@
1397
  "model.layers.22.self_attn.v_proj.codebooks": "model-00002-of-00003.safetensors",
1398
  "model.layers.22.self_attn.v_proj.codes": "model-00002-of-00003.safetensors",
1399
  "model.layers.22.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
1400
- "model.layers.23.block_sparse_moe.experts.0.w1.codebooks": "model-00003-of-00003.safetensors",
1401
- "model.layers.23.block_sparse_moe.experts.0.w1.codes": "model-00003-of-00003.safetensors",
1402
- "model.layers.23.block_sparse_moe.experts.0.w1.scales": "model-00003-of-00003.safetensors",
1403
- "model.layers.23.block_sparse_moe.experts.0.w2.codebooks": "model-00003-of-00003.safetensors",
1404
- "model.layers.23.block_sparse_moe.experts.0.w2.codes": "model-00003-of-00003.safetensors",
1405
- "model.layers.23.block_sparse_moe.experts.0.w2.scales": "model-00003-of-00003.safetensors",
1406
- "model.layers.23.block_sparse_moe.experts.0.w3.codebooks": "model-00003-of-00003.safetensors",
1407
- "model.layers.23.block_sparse_moe.experts.0.w3.codes": "model-00003-of-00003.safetensors",
1408
- "model.layers.23.block_sparse_moe.experts.0.w3.scales": "model-00003-of-00003.safetensors",
1409
- "model.layers.23.block_sparse_moe.experts.1.w1.codebooks": "model-00003-of-00003.safetensors",
1410
- "model.layers.23.block_sparse_moe.experts.1.w1.codes": "model-00003-of-00003.safetensors",
1411
- "model.layers.23.block_sparse_moe.experts.1.w1.scales": "model-00003-of-00003.safetensors",
1412
- "model.layers.23.block_sparse_moe.experts.1.w2.codebooks": "model-00003-of-00003.safetensors",
1413
- "model.layers.23.block_sparse_moe.experts.1.w2.codes": "model-00003-of-00003.safetensors",
1414
- "model.layers.23.block_sparse_moe.experts.1.w2.scales": "model-00003-of-00003.safetensors",
1415
- "model.layers.23.block_sparse_moe.experts.1.w3.codebooks": "model-00003-of-00003.safetensors",
1416
- "model.layers.23.block_sparse_moe.experts.1.w3.codes": "model-00003-of-00003.safetensors",
1417
- "model.layers.23.block_sparse_moe.experts.1.w3.scales": "model-00003-of-00003.safetensors",
1418
- "model.layers.23.block_sparse_moe.experts.2.w1.codebooks": "model-00003-of-00003.safetensors",
1419
- "model.layers.23.block_sparse_moe.experts.2.w1.codes": "model-00003-of-00003.safetensors",
1420
- "model.layers.23.block_sparse_moe.experts.2.w1.scales": "model-00003-of-00003.safetensors",
1421
- "model.layers.23.block_sparse_moe.experts.2.w2.codebooks": "model-00003-of-00003.safetensors",
1422
- "model.layers.23.block_sparse_moe.experts.2.w2.codes": "model-00003-of-00003.safetensors",
1423
- "model.layers.23.block_sparse_moe.experts.2.w2.scales": "model-00003-of-00003.safetensors",
1424
- "model.layers.23.block_sparse_moe.experts.2.w3.codebooks": "model-00003-of-00003.safetensors",
1425
- "model.layers.23.block_sparse_moe.experts.2.w3.codes": "model-00003-of-00003.safetensors",
1426
- "model.layers.23.block_sparse_moe.experts.2.w3.scales": "model-00003-of-00003.safetensors",
1427
- "model.layers.23.block_sparse_moe.experts.3.w1.codebooks": "model-00003-of-00003.safetensors",
1428
- "model.layers.23.block_sparse_moe.experts.3.w1.codes": "model-00003-of-00003.safetensors",
1429
- "model.layers.23.block_sparse_moe.experts.3.w1.scales": "model-00003-of-00003.safetensors",
1430
- "model.layers.23.block_sparse_moe.experts.3.w2.codebooks": "model-00003-of-00003.safetensors",
1431
- "model.layers.23.block_sparse_moe.experts.3.w2.codes": "model-00003-of-00003.safetensors",
1432
- "model.layers.23.block_sparse_moe.experts.3.w2.scales": "model-00003-of-00003.safetensors",
1433
- "model.layers.23.block_sparse_moe.experts.3.w3.codebooks": "model-00003-of-00003.safetensors",
1434
- "model.layers.23.block_sparse_moe.experts.3.w3.codes": "model-00003-of-00003.safetensors",
1435
- "model.layers.23.block_sparse_moe.experts.3.w3.scales": "model-00003-of-00003.safetensors",
1436
- "model.layers.23.block_sparse_moe.experts.4.w1.codebooks": "model-00003-of-00003.safetensors",
1437
- "model.layers.23.block_sparse_moe.experts.4.w1.codes": "model-00003-of-00003.safetensors",
1438
- "model.layers.23.block_sparse_moe.experts.4.w1.scales": "model-00003-of-00003.safetensors",
1439
- "model.layers.23.block_sparse_moe.experts.4.w2.codebooks": "model-00003-of-00003.safetensors",
1440
- "model.layers.23.block_sparse_moe.experts.4.w2.codes": "model-00003-of-00003.safetensors",
1441
- "model.layers.23.block_sparse_moe.experts.4.w2.scales": "model-00003-of-00003.safetensors",
1442
- "model.layers.23.block_sparse_moe.experts.4.w3.codebooks": "model-00003-of-00003.safetensors",
1443
- "model.layers.23.block_sparse_moe.experts.4.w3.codes": "model-00003-of-00003.safetensors",
1444
- "model.layers.23.block_sparse_moe.experts.4.w3.scales": "model-00003-of-00003.safetensors",
1445
- "model.layers.23.block_sparse_moe.experts.5.w1.codebooks": "model-00003-of-00003.safetensors",
1446
- "model.layers.23.block_sparse_moe.experts.5.w1.codes": "model-00003-of-00003.safetensors",
1447
- "model.layers.23.block_sparse_moe.experts.5.w1.scales": "model-00003-of-00003.safetensors",
1448
- "model.layers.23.block_sparse_moe.experts.5.w2.codebooks": "model-00003-of-00003.safetensors",
1449
- "model.layers.23.block_sparse_moe.experts.5.w2.codes": "model-00003-of-00003.safetensors",
1450
- "model.layers.23.block_sparse_moe.experts.5.w2.scales": "model-00003-of-00003.safetensors",
1451
- "model.layers.23.block_sparse_moe.experts.5.w3.codebooks": "model-00003-of-00003.safetensors",
1452
- "model.layers.23.block_sparse_moe.experts.5.w3.codes": "model-00003-of-00003.safetensors",
1453
- "model.layers.23.block_sparse_moe.experts.5.w3.scales": "model-00003-of-00003.safetensors",
1454
- "model.layers.23.block_sparse_moe.experts.6.w1.codebooks": "model-00003-of-00003.safetensors",
1455
- "model.layers.23.block_sparse_moe.experts.6.w1.codes": "model-00003-of-00003.safetensors",
1456
- "model.layers.23.block_sparse_moe.experts.6.w1.scales": "model-00003-of-00003.safetensors",
1457
- "model.layers.23.block_sparse_moe.experts.6.w2.codebooks": "model-00003-of-00003.safetensors",
1458
- "model.layers.23.block_sparse_moe.experts.6.w2.codes": "model-00003-of-00003.safetensors",
1459
- "model.layers.23.block_sparse_moe.experts.6.w2.scales": "model-00003-of-00003.safetensors",
1460
- "model.layers.23.block_sparse_moe.experts.6.w3.codebooks": "model-00003-of-00003.safetensors",
1461
- "model.layers.23.block_sparse_moe.experts.6.w3.codes": "model-00003-of-00003.safetensors",
1462
- "model.layers.23.block_sparse_moe.experts.6.w3.scales": "model-00003-of-00003.safetensors",
1463
- "model.layers.23.block_sparse_moe.experts.7.w1.codebooks": "model-00003-of-00003.safetensors",
1464
- "model.layers.23.block_sparse_moe.experts.7.w1.codes": "model-00003-of-00003.safetensors",
1465
- "model.layers.23.block_sparse_moe.experts.7.w1.scales": "model-00003-of-00003.safetensors",
1466
- "model.layers.23.block_sparse_moe.experts.7.w2.codebooks": "model-00003-of-00003.safetensors",
1467
- "model.layers.23.block_sparse_moe.experts.7.w2.codes": "model-00003-of-00003.safetensors",
1468
- "model.layers.23.block_sparse_moe.experts.7.w2.scales": "model-00003-of-00003.safetensors",
1469
- "model.layers.23.block_sparse_moe.experts.7.w3.codebooks": "model-00003-of-00003.safetensors",
1470
- "model.layers.23.block_sparse_moe.experts.7.w3.codes": "model-00003-of-00003.safetensors",
1471
- "model.layers.23.block_sparse_moe.experts.7.w3.scales": "model-00003-of-00003.safetensors",
1472
- "model.layers.23.block_sparse_moe.gate.weight": "model-00003-of-00003.safetensors",
1473
- "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
1474
- "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
1475
- "model.layers.23.self_attn.k_proj.codebooks": "model-00003-of-00003.safetensors",
1476
- "model.layers.23.self_attn.k_proj.codes": "model-00003-of-00003.safetensors",
1477
- "model.layers.23.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
1478
- "model.layers.23.self_attn.o_proj.codebooks": "model-00003-of-00003.safetensors",
1479
- "model.layers.23.self_attn.o_proj.codes": "model-00003-of-00003.safetensors",
1480
- "model.layers.23.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
1481
- "model.layers.23.self_attn.q_proj.codebooks": "model-00003-of-00003.safetensors",
1482
- "model.layers.23.self_attn.q_proj.codes": "model-00003-of-00003.safetensors",
1483
- "model.layers.23.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
1484
- "model.layers.23.self_attn.v_proj.codebooks": "model-00003-of-00003.safetensors",
1485
- "model.layers.23.self_attn.v_proj.codes": "model-00003-of-00003.safetensors",
1486
- "model.layers.23.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
1487
- "model.layers.24.block_sparse_moe.experts.0.w1.codebooks": "model-00003-of-00003.safetensors",
1488
- "model.layers.24.block_sparse_moe.experts.0.w1.codes": "model-00003-of-00003.safetensors",
1489
- "model.layers.24.block_sparse_moe.experts.0.w1.scales": "model-00003-of-00003.safetensors",
1490
- "model.layers.24.block_sparse_moe.experts.0.w2.codebooks": "model-00003-of-00003.safetensors",
1491
- "model.layers.24.block_sparse_moe.experts.0.w2.codes": "model-00003-of-00003.safetensors",
1492
- "model.layers.24.block_sparse_moe.experts.0.w2.scales": "model-00003-of-00003.safetensors",
1493
- "model.layers.24.block_sparse_moe.experts.0.w3.codebooks": "model-00003-of-00003.safetensors",
1494
- "model.layers.24.block_sparse_moe.experts.0.w3.codes": "model-00003-of-00003.safetensors",
1495
- "model.layers.24.block_sparse_moe.experts.0.w3.scales": "model-00003-of-00003.safetensors",
1496
- "model.layers.24.block_sparse_moe.experts.1.w1.codebooks": "model-00003-of-00003.safetensors",
1497
- "model.layers.24.block_sparse_moe.experts.1.w1.codes": "model-00003-of-00003.safetensors",
1498
- "model.layers.24.block_sparse_moe.experts.1.w1.scales": "model-00003-of-00003.safetensors",
1499
- "model.layers.24.block_sparse_moe.experts.1.w2.codebooks": "model-00003-of-00003.safetensors",
1500
- "model.layers.24.block_sparse_moe.experts.1.w2.codes": "model-00003-of-00003.safetensors",
1501
- "model.layers.24.block_sparse_moe.experts.1.w2.scales": "model-00003-of-00003.safetensors",
1502
- "model.layers.24.block_sparse_moe.experts.1.w3.codebooks": "model-00003-of-00003.safetensors",
1503
- "model.layers.24.block_sparse_moe.experts.1.w3.codes": "model-00003-of-00003.safetensors",
1504
- "model.layers.24.block_sparse_moe.experts.1.w3.scales": "model-00003-of-00003.safetensors",
1505
- "model.layers.24.block_sparse_moe.experts.2.w1.codebooks": "model-00003-of-00003.safetensors",
1506
- "model.layers.24.block_sparse_moe.experts.2.w1.codes": "model-00003-of-00003.safetensors",
1507
- "model.layers.24.block_sparse_moe.experts.2.w1.scales": "model-00003-of-00003.safetensors",
1508
- "model.layers.24.block_sparse_moe.experts.2.w2.codebooks": "model-00003-of-00003.safetensors",
1509
- "model.layers.24.block_sparse_moe.experts.2.w2.codes": "model-00003-of-00003.safetensors",
1510
- "model.layers.24.block_sparse_moe.experts.2.w2.scales": "model-00003-of-00003.safetensors",
1511
- "model.layers.24.block_sparse_moe.experts.2.w3.codebooks": "model-00003-of-00003.safetensors",
1512
- "model.layers.24.block_sparse_moe.experts.2.w3.codes": "model-00003-of-00003.safetensors",
1513
- "model.layers.24.block_sparse_moe.experts.2.w3.scales": "model-00003-of-00003.safetensors",
1514
- "model.layers.24.block_sparse_moe.experts.3.w1.codebooks": "model-00003-of-00003.safetensors",
1515
- "model.layers.24.block_sparse_moe.experts.3.w1.codes": "model-00003-of-00003.safetensors",
1516
- "model.layers.24.block_sparse_moe.experts.3.w1.scales": "model-00003-of-00003.safetensors",
1517
- "model.layers.24.block_sparse_moe.experts.3.w2.codebooks": "model-00003-of-00003.safetensors",
1518
- "model.layers.24.block_sparse_moe.experts.3.w2.codes": "model-00003-of-00003.safetensors",
1519
- "model.layers.24.block_sparse_moe.experts.3.w2.scales": "model-00003-of-00003.safetensors",
1520
- "model.layers.24.block_sparse_moe.experts.3.w3.codebooks": "model-00003-of-00003.safetensors",
1521
- "model.layers.24.block_sparse_moe.experts.3.w3.codes": "model-00003-of-00003.safetensors",
1522
- "model.layers.24.block_sparse_moe.experts.3.w3.scales": "model-00003-of-00003.safetensors",
1523
- "model.layers.24.block_sparse_moe.experts.4.w1.codebooks": "model-00003-of-00003.safetensors",
1524
- "model.layers.24.block_sparse_moe.experts.4.w1.codes": "model-00003-of-00003.safetensors",
1525
- "model.layers.24.block_sparse_moe.experts.4.w1.scales": "model-00003-of-00003.safetensors",
1526
- "model.layers.24.block_sparse_moe.experts.4.w2.codebooks": "model-00003-of-00003.safetensors",
1527
- "model.layers.24.block_sparse_moe.experts.4.w2.codes": "model-00003-of-00003.safetensors",
1528
- "model.layers.24.block_sparse_moe.experts.4.w2.scales": "model-00003-of-00003.safetensors",
1529
- "model.layers.24.block_sparse_moe.experts.4.w3.codebooks": "model-00003-of-00003.safetensors",
1530
- "model.layers.24.block_sparse_moe.experts.4.w3.codes": "model-00003-of-00003.safetensors",
1531
- "model.layers.24.block_sparse_moe.experts.4.w3.scales": "model-00003-of-00003.safetensors",
1532
- "model.layers.24.block_sparse_moe.experts.5.w1.codebooks": "model-00003-of-00003.safetensors",
1533
- "model.layers.24.block_sparse_moe.experts.5.w1.codes": "model-00003-of-00003.safetensors",
1534
- "model.layers.24.block_sparse_moe.experts.5.w1.scales": "model-00003-of-00003.safetensors",
1535
- "model.layers.24.block_sparse_moe.experts.5.w2.codebooks": "model-00003-of-00003.safetensors",
1536
- "model.layers.24.block_sparse_moe.experts.5.w2.codes": "model-00003-of-00003.safetensors",
1537
- "model.layers.24.block_sparse_moe.experts.5.w2.scales": "model-00003-of-00003.safetensors",
1538
- "model.layers.24.block_sparse_moe.experts.5.w3.codebooks": "model-00003-of-00003.safetensors",
1539
- "model.layers.24.block_sparse_moe.experts.5.w3.codes": "model-00003-of-00003.safetensors",
1540
- "model.layers.24.block_sparse_moe.experts.5.w3.scales": "model-00003-of-00003.safetensors",
1541
- "model.layers.24.block_sparse_moe.experts.6.w1.codebooks": "model-00003-of-00003.safetensors",
1542
  "model.layers.24.block_sparse_moe.experts.6.w1.codes": "model-00003-of-00003.safetensors",
1543
  "model.layers.24.block_sparse_moe.experts.6.w1.scales": "model-00003-of-00003.safetensors",
1544
  "model.layers.24.block_sparse_moe.experts.6.w2.codebooks": "model-00003-of-00003.safetensors",
@@ -1556,21 +1556,21 @@
1556
  "model.layers.24.block_sparse_moe.experts.7.w3.codebooks": "model-00003-of-00003.safetensors",
1557
  "model.layers.24.block_sparse_moe.experts.7.w3.codes": "model-00003-of-00003.safetensors",
1558
  "model.layers.24.block_sparse_moe.experts.7.w3.scales": "model-00003-of-00003.safetensors",
1559
- "model.layers.24.block_sparse_moe.gate.weight": "model-00003-of-00003.safetensors",
1560
  "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
1561
  "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
1562
- "model.layers.24.self_attn.k_proj.codebooks": "model-00003-of-00003.safetensors",
1563
- "model.layers.24.self_attn.k_proj.codes": "model-00003-of-00003.safetensors",
1564
- "model.layers.24.self_attn.k_proj.scales": "model-00003-of-00003.safetensors",
1565
- "model.layers.24.self_attn.o_proj.codebooks": "model-00003-of-00003.safetensors",
1566
- "model.layers.24.self_attn.o_proj.codes": "model-00003-of-00003.safetensors",
1567
- "model.layers.24.self_attn.o_proj.scales": "model-00003-of-00003.safetensors",
1568
- "model.layers.24.self_attn.q_proj.codebooks": "model-00003-of-00003.safetensors",
1569
- "model.layers.24.self_attn.q_proj.codes": "model-00003-of-00003.safetensors",
1570
- "model.layers.24.self_attn.q_proj.scales": "model-00003-of-00003.safetensors",
1571
- "model.layers.24.self_attn.v_proj.codebooks": "model-00003-of-00003.safetensors",
1572
- "model.layers.24.self_attn.v_proj.codes": "model-00003-of-00003.safetensors",
1573
- "model.layers.24.self_attn.v_proj.scales": "model-00003-of-00003.safetensors",
1574
  "model.layers.25.block_sparse_moe.experts.0.w1.codebooks": "model-00003-of-00003.safetensors",
1575
  "model.layers.25.block_sparse_moe.experts.0.w1.codes": "model-00003-of-00003.safetensors",
1576
  "model.layers.25.block_sparse_moe.experts.0.w1.scales": "model-00003-of-00003.safetensors",
 
1
  {
2
  "metadata": {
3
+ "total_size": 13093707776
4
  },
5
  "weight_map": {
6
  "lm_head.weight": "model-00003-of-00003.safetensors",
 
219
  "model.layers.10.block_sparse_moe.experts.4.w1.codes": "model-00001-of-00003.safetensors",
220
  "model.layers.10.block_sparse_moe.experts.4.w1.scales": "model-00001-of-00003.safetensors",
221
  "model.layers.10.block_sparse_moe.experts.4.w2.codebooks": "model-00001-of-00003.safetensors",
222
+ "model.layers.10.block_sparse_moe.experts.4.w2.codes": "model-00001-of-00003.safetensors",
223
+ "model.layers.10.block_sparse_moe.experts.4.w2.scales": "model-00001-of-00003.safetensors",
224
+ "model.layers.10.block_sparse_moe.experts.4.w3.codebooks": "model-00001-of-00003.safetensors",
225
+ "model.layers.10.block_sparse_moe.experts.4.w3.codes": "model-00001-of-00003.safetensors",
226
+ "model.layers.10.block_sparse_moe.experts.4.w3.scales": "model-00001-of-00003.safetensors",
227
+ "model.layers.10.block_sparse_moe.experts.5.w1.codebooks": "model-00001-of-00003.safetensors",
228
+ "model.layers.10.block_sparse_moe.experts.5.w1.codes": "model-00001-of-00003.safetensors",
229
+ "model.layers.10.block_sparse_moe.experts.5.w1.scales": "model-00001-of-00003.safetensors",
230
+ "model.layers.10.block_sparse_moe.experts.5.w2.codebooks": "model-00001-of-00003.safetensors",
231
+ "model.layers.10.block_sparse_moe.experts.5.w2.codes": "model-00001-of-00003.safetensors",
232
+ "model.layers.10.block_sparse_moe.experts.5.w2.scales": "model-00001-of-00003.safetensors",
233
+ "model.layers.10.block_sparse_moe.experts.5.w3.codebooks": "model-00001-of-00003.safetensors",
234
+ "model.layers.10.block_sparse_moe.experts.5.w3.codes": "model-00001-of-00003.safetensors",
235
+ "model.layers.10.block_sparse_moe.experts.5.w3.scales": "model-00001-of-00003.safetensors",
236
+ "model.layers.10.block_sparse_moe.experts.6.w1.codebooks": "model-00001-of-00003.safetensors",
237
+ "model.layers.10.block_sparse_moe.experts.6.w1.codes": "model-00001-of-00003.safetensors",
238
+ "model.layers.10.block_sparse_moe.experts.6.w1.scales": "model-00001-of-00003.safetensors",
239
+ "model.layers.10.block_sparse_moe.experts.6.w2.codebooks": "model-00001-of-00003.safetensors",
240
+ "model.layers.10.block_sparse_moe.experts.6.w2.codes": "model-00001-of-00003.safetensors",
241
+ "model.layers.10.block_sparse_moe.experts.6.w2.scales": "model-00001-of-00003.safetensors",
242
+ "model.layers.10.block_sparse_moe.experts.6.w3.codebooks": "model-00001-of-00003.safetensors",
243
+ "model.layers.10.block_sparse_moe.experts.6.w3.codes": "model-00001-of-00003.safetensors",
244
+ "model.layers.10.block_sparse_moe.experts.6.w3.scales": "model-00001-of-00003.safetensors",
245
+ "model.layers.10.block_sparse_moe.experts.7.w1.codebooks": "model-00001-of-00003.safetensors",
246
+ "model.layers.10.block_sparse_moe.experts.7.w1.codes": "model-00001-of-00003.safetensors",
247
+ "model.layers.10.block_sparse_moe.experts.7.w1.scales": "model-00001-of-00003.safetensors",
248
+ "model.layers.10.block_sparse_moe.experts.7.w2.codebooks": "model-00001-of-00003.safetensors",
249
+ "model.layers.10.block_sparse_moe.experts.7.w2.codes": "model-00001-of-00003.safetensors",
250
+ "model.layers.10.block_sparse_moe.experts.7.w2.scales": "model-00001-of-00003.safetensors",
251
+ "model.layers.10.block_sparse_moe.experts.7.w3.codebooks": "model-00001-of-00003.safetensors",
252
+ "model.layers.10.block_sparse_moe.experts.7.w3.codes": "model-00001-of-00003.safetensors",
253
+ "model.layers.10.block_sparse_moe.experts.7.w3.scales": "model-00001-of-00003.safetensors",
254
  "model.layers.10.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
255
+ "model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
256
+ "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
257
  "model.layers.10.self_attn.k_proj.codebooks": "model-00001-of-00003.safetensors",
258
  "model.layers.10.self_attn.k_proj.codes": "model-00001-of-00003.safetensors",
259
  "model.layers.10.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
 
266
  "model.layers.10.self_attn.v_proj.codebooks": "model-00001-of-00003.safetensors",
267
  "model.layers.10.self_attn.v_proj.codes": "model-00001-of-00003.safetensors",
268
  "model.layers.10.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
269
+ "model.layers.11.block_sparse_moe.experts.0.w1.codebooks": "model-00001-of-00003.safetensors",
270
+ "model.layers.11.block_sparse_moe.experts.0.w1.codes": "model-00001-of-00003.safetensors",
271
+ "model.layers.11.block_sparse_moe.experts.0.w1.scales": "model-00001-of-00003.safetensors",
272
+ "model.layers.11.block_sparse_moe.experts.0.w2.codebooks": "model-00001-of-00003.safetensors",
273
+ "model.layers.11.block_sparse_moe.experts.0.w2.codes": "model-00001-of-00003.safetensors",
274
+ "model.layers.11.block_sparse_moe.experts.0.w2.scales": "model-00001-of-00003.safetensors",
275
+ "model.layers.11.block_sparse_moe.experts.0.w3.codebooks": "model-00001-of-00003.safetensors",
276
+ "model.layers.11.block_sparse_moe.experts.0.w3.codes": "model-00001-of-00003.safetensors",
277
+ "model.layers.11.block_sparse_moe.experts.0.w3.scales": "model-00001-of-00003.safetensors",
278
+ "model.layers.11.block_sparse_moe.experts.1.w1.codebooks": "model-00001-of-00003.safetensors",
279
+ "model.layers.11.block_sparse_moe.experts.1.w1.codes": "model-00001-of-00003.safetensors",
280
+ "model.layers.11.block_sparse_moe.experts.1.w1.scales": "model-00001-of-00003.safetensors",
281
+ "model.layers.11.block_sparse_moe.experts.1.w2.codebooks": "model-00001-of-00003.safetensors",
282
+ "model.layers.11.block_sparse_moe.experts.1.w2.codes": "model-00001-of-00003.safetensors",
283
+ "model.layers.11.block_sparse_moe.experts.1.w2.scales": "model-00001-of-00003.safetensors",
284
+ "model.layers.11.block_sparse_moe.experts.1.w3.codebooks": "model-00001-of-00003.safetensors",
285
+ "model.layers.11.block_sparse_moe.experts.1.w3.codes": "model-00001-of-00003.safetensors",
286
+ "model.layers.11.block_sparse_moe.experts.1.w3.scales": "model-00001-of-00003.safetensors",
287
+ "model.layers.11.block_sparse_moe.experts.2.w1.codebooks": "model-00001-of-00003.safetensors",
288
+ "model.layers.11.block_sparse_moe.experts.2.w1.codes": "model-00001-of-00003.safetensors",
289
+ "model.layers.11.block_sparse_moe.experts.2.w1.scales": "model-00001-of-00003.safetensors",
290
+ "model.layers.11.block_sparse_moe.experts.2.w2.codebooks": "model-00001-of-00003.safetensors",
291
+ "model.layers.11.block_sparse_moe.experts.2.w2.codes": "model-00001-of-00003.safetensors",
292
+ "model.layers.11.block_sparse_moe.experts.2.w2.scales": "model-00001-of-00003.safetensors",
293
+ "model.layers.11.block_sparse_moe.experts.2.w3.codebooks": "model-00001-of-00003.safetensors",
294
+ "model.layers.11.block_sparse_moe.experts.2.w3.codes": "model-00001-of-00003.safetensors",
295
+ "model.layers.11.block_sparse_moe.experts.2.w3.scales": "model-00001-of-00003.safetensors",
296
+ "model.layers.11.block_sparse_moe.experts.3.w1.codebooks": "model-00001-of-00003.safetensors",
297
+ "model.layers.11.block_sparse_moe.experts.3.w1.codes": "model-00001-of-00003.safetensors",
298
+ "model.layers.11.block_sparse_moe.experts.3.w1.scales": "model-00001-of-00003.safetensors",
299
+ "model.layers.11.block_sparse_moe.experts.3.w2.codebooks": "model-00001-of-00003.safetensors",
300
+ "model.layers.11.block_sparse_moe.experts.3.w2.codes": "model-00001-of-00003.safetensors",
301
+ "model.layers.11.block_sparse_moe.experts.3.w2.scales": "model-00001-of-00003.safetensors",
302
+ "model.layers.11.block_sparse_moe.experts.3.w3.codebooks": "model-00001-of-00003.safetensors",
303
+ "model.layers.11.block_sparse_moe.experts.3.w3.codes": "model-00001-of-00003.safetensors",
304
+ "model.layers.11.block_sparse_moe.experts.3.w3.scales": "model-00001-of-00003.safetensors",
305
+ "model.layers.11.block_sparse_moe.experts.4.w1.codebooks": "model-00001-of-00003.safetensors",
306
+ "model.layers.11.block_sparse_moe.experts.4.w1.codes": "model-00001-of-00003.safetensors",
307
+ "model.layers.11.block_sparse_moe.experts.4.w1.scales": "model-00001-of-00003.safetensors",
308
+ "model.layers.11.block_sparse_moe.experts.4.w2.codebooks": "model-00001-of-00003.safetensors",
309
+ "model.layers.11.block_sparse_moe.experts.4.w2.codes": "model-00001-of-00003.safetensors",
310
+ "model.layers.11.block_sparse_moe.experts.4.w2.scales": "model-00001-of-00003.safetensors",
311
+ "model.layers.11.block_sparse_moe.experts.4.w3.codebooks": "model-00001-of-00003.safetensors",
312
+ "model.layers.11.block_sparse_moe.experts.4.w3.codes": "model-00001-of-00003.safetensors",
313
+ "model.layers.11.block_sparse_moe.experts.4.w3.scales": "model-00001-of-00003.safetensors",
314
+ "model.layers.11.block_sparse_moe.experts.5.w1.codebooks": "model-00001-of-00003.safetensors",
315
+ "model.layers.11.block_sparse_moe.experts.5.w1.codes": "model-00001-of-00003.safetensors",
316
+ "model.layers.11.block_sparse_moe.experts.5.w1.scales": "model-00001-of-00003.safetensors",
317
+ "model.layers.11.block_sparse_moe.experts.5.w2.codebooks": "model-00001-of-00003.safetensors",
318
+ "model.layers.11.block_sparse_moe.experts.5.w2.codes": "model-00001-of-00003.safetensors",
319
+ "model.layers.11.block_sparse_moe.experts.5.w2.scales": "model-00001-of-00003.safetensors",
320
+ "model.layers.11.block_sparse_moe.experts.5.w3.codebooks": "model-00001-of-00003.safetensors",
321
+ "model.layers.11.block_sparse_moe.experts.5.w3.codes": "model-00001-of-00003.safetensors",
322
+ "model.layers.11.block_sparse_moe.experts.5.w3.scales": "model-00001-of-00003.safetensors",
323
+ "model.layers.11.block_sparse_moe.experts.6.w1.codebooks": "model-00001-of-00003.safetensors",
324
+ "model.layers.11.block_sparse_moe.experts.6.w1.codes": "model-00001-of-00003.safetensors",
325
+ "model.layers.11.block_sparse_moe.experts.6.w1.scales": "model-00001-of-00003.safetensors",
326
+ "model.layers.11.block_sparse_moe.experts.6.w2.codebooks": "model-00001-of-00003.safetensors",
327
+ "model.layers.11.block_sparse_moe.experts.6.w2.codes": "model-00001-of-00003.safetensors",
328
+ "model.layers.11.block_sparse_moe.experts.6.w2.scales": "model-00001-of-00003.safetensors",
329
+ "model.layers.11.block_sparse_moe.experts.6.w3.codebooks": "model-00001-of-00003.safetensors",
330
+ "model.layers.11.block_sparse_moe.experts.6.w3.codes": "model-00001-of-00003.safetensors",
331
+ "model.layers.11.block_sparse_moe.experts.6.w3.scales": "model-00001-of-00003.safetensors",
332
+ "model.layers.11.block_sparse_moe.experts.7.w1.codebooks": "model-00001-of-00003.safetensors",
333
+ "model.layers.11.block_sparse_moe.experts.7.w1.codes": "model-00001-of-00003.safetensors",
334
+ "model.layers.11.block_sparse_moe.experts.7.w1.scales": "model-00001-of-00003.safetensors",
335
+ "model.layers.11.block_sparse_moe.experts.7.w2.codebooks": "model-00001-of-00003.safetensors",
336
+ "model.layers.11.block_sparse_moe.experts.7.w2.codes": "model-00001-of-00003.safetensors",
337
+ "model.layers.11.block_sparse_moe.experts.7.w2.scales": "model-00001-of-00003.safetensors",
338
+ "model.layers.11.block_sparse_moe.experts.7.w3.codebooks": "model-00001-of-00003.safetensors",
339
+ "model.layers.11.block_sparse_moe.experts.7.w3.codes": "model-00001-of-00003.safetensors",
340
+ "model.layers.11.block_sparse_moe.experts.7.w3.scales": "model-00001-of-00003.safetensors",
341
+ "model.layers.11.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
342
+ "model.layers.11.input_layernorm.weight": "model-00001-of-00003.safetensors",
343
+ "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
344
+ "model.layers.11.self_attn.k_proj.codebooks": "model-00001-of-00003.safetensors",
345
+ "model.layers.11.self_attn.k_proj.codes": "model-00001-of-00003.safetensors",
346
+ "model.layers.11.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
347
+ "model.layers.11.self_attn.o_proj.codebooks": "model-00001-of-00003.safetensors",
348
+ "model.layers.11.self_attn.o_proj.codes": "model-00001-of-00003.safetensors",
349
+ "model.layers.11.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
350
+ "model.layers.11.self_attn.q_proj.codebooks": "model-00001-of-00003.safetensors",
351
+ "model.layers.11.self_attn.q_proj.codes": "model-00001-of-00003.safetensors",
352
+ "model.layers.11.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
353
+ "model.layers.11.self_attn.v_proj.codebooks": "model-00001-of-00003.safetensors",
354
+ "model.layers.11.self_attn.v_proj.codes": "model-00001-of-00003.safetensors",
355
+ "model.layers.11.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
356
+ "model.layers.12.block_sparse_moe.experts.0.w1.codebooks": "model-00001-of-00003.safetensors",
357
  "model.layers.12.block_sparse_moe.experts.0.w1.codes": "model-00002-of-00003.safetensors",
358
  "model.layers.12.block_sparse_moe.experts.0.w1.scales": "model-00002-of-00003.safetensors",
359
  "model.layers.12.block_sparse_moe.experts.0.w2.codebooks": "model-00002-of-00003.safetensors",
 
425
  "model.layers.12.block_sparse_moe.experts.7.w3.codebooks": "model-00002-of-00003.safetensors",
426
  "model.layers.12.block_sparse_moe.experts.7.w3.codes": "model-00002-of-00003.safetensors",
427
  "model.layers.12.block_sparse_moe.experts.7.w3.scales": "model-00002-of-00003.safetensors",
428
+ "model.layers.12.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
429
  "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
430
  "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
431
+ "model.layers.12.self_attn.k_proj.codebooks": "model-00001-of-00003.safetensors",
432
+ "model.layers.12.self_attn.k_proj.codes": "model-00001-of-00003.safetensors",
433
+ "model.layers.12.self_attn.k_proj.scales": "model-00001-of-00003.safetensors",
434
+ "model.layers.12.self_attn.o_proj.codebooks": "model-00001-of-00003.safetensors",
435
+ "model.layers.12.self_attn.o_proj.codes": "model-00001-of-00003.safetensors",
436
+ "model.layers.12.self_attn.o_proj.scales": "model-00001-of-00003.safetensors",
437
+ "model.layers.12.self_attn.q_proj.codebooks": "model-00001-of-00003.safetensors",
438
+ "model.layers.12.self_attn.q_proj.codes": "model-00001-of-00003.safetensors",
439
+ "model.layers.12.self_attn.q_proj.scales": "model-00001-of-00003.safetensors",
440
+ "model.layers.12.self_attn.v_proj.codebooks": "model-00001-of-00003.safetensors",
441
+ "model.layers.12.self_attn.v_proj.codes": "model-00001-of-00003.safetensors",
442
+ "model.layers.12.self_attn.v_proj.scales": "model-00001-of-00003.safetensors",
443
  "model.layers.13.block_sparse_moe.experts.0.w1.codebooks": "model-00002-of-00003.safetensors",
444
  "model.layers.13.block_sparse_moe.experts.0.w1.codes": "model-00002-of-00003.safetensors",
445
  "model.layers.13.block_sparse_moe.experts.0.w1.scales": "model-00002-of-00003.safetensors",
 
1335
  "model.layers.22.block_sparse_moe.experts.2.w2.codes": "model-00002-of-00003.safetensors",
1336
  "model.layers.22.block_sparse_moe.experts.2.w2.scales": "model-00002-of-00003.safetensors",
1337
  "model.layers.22.block_sparse_moe.experts.2.w3.codebooks": "model-00002-of-00003.safetensors",
1338
+ "model.layers.22.block_sparse_moe.experts.2.w3.codes": "model-00002-of-00003.safetensors",
1339
+ "model.layers.22.block_sparse_moe.experts.2.w3.scales": "model-00002-of-00003.safetensors",
1340
+ "model.layers.22.block_sparse_moe.experts.3.w1.codebooks": "model-00002-of-00003.safetensors",
1341
+ "model.layers.22.block_sparse_moe.experts.3.w1.codes": "model-00002-of-00003.safetensors",
1342
+ "model.layers.22.block_sparse_moe.experts.3.w1.scales": "model-00002-of-00003.safetensors",
1343
+ "model.layers.22.block_sparse_moe.experts.3.w2.codebooks": "model-00002-of-00003.safetensors",
1344
+ "model.layers.22.block_sparse_moe.experts.3.w2.codes": "model-00002-of-00003.safetensors",
1345
+ "model.layers.22.block_sparse_moe.experts.3.w2.scales": "model-00002-of-00003.safetensors",
1346
+ "model.layers.22.block_sparse_moe.experts.3.w3.codebooks": "model-00002-of-00003.safetensors",
1347
+ "model.layers.22.block_sparse_moe.experts.3.w3.codes": "model-00002-of-00003.safetensors",
1348
+ "model.layers.22.block_sparse_moe.experts.3.w3.scales": "model-00002-of-00003.safetensors",
1349
+ "model.layers.22.block_sparse_moe.experts.4.w1.codebooks": "model-00002-of-00003.safetensors",
1350
+ "model.layers.22.block_sparse_moe.experts.4.w1.codes": "model-00002-of-00003.safetensors",
1351
+ "model.layers.22.block_sparse_moe.experts.4.w1.scales": "model-00002-of-00003.safetensors",
1352
+ "model.layers.22.block_sparse_moe.experts.4.w2.codebooks": "model-00002-of-00003.safetensors",
1353
+ "model.layers.22.block_sparse_moe.experts.4.w2.codes": "model-00002-of-00003.safetensors",
1354
+ "model.layers.22.block_sparse_moe.experts.4.w2.scales": "model-00002-of-00003.safetensors",
1355
+ "model.layers.22.block_sparse_moe.experts.4.w3.codebooks": "model-00002-of-00003.safetensors",
1356
+ "model.layers.22.block_sparse_moe.experts.4.w3.codes": "model-00002-of-00003.safetensors",
1357
+ "model.layers.22.block_sparse_moe.experts.4.w3.scales": "model-00002-of-00003.safetensors",
1358
+ "model.layers.22.block_sparse_moe.experts.5.w1.codebooks": "model-00002-of-00003.safetensors",
1359
+ "model.layers.22.block_sparse_moe.experts.5.w1.codes": "model-00002-of-00003.safetensors",
1360
+ "model.layers.22.block_sparse_moe.experts.5.w1.scales": "model-00002-of-00003.safetensors",
1361
+ "model.layers.22.block_sparse_moe.experts.5.w2.codebooks": "model-00002-of-00003.safetensors",
1362
+ "model.layers.22.block_sparse_moe.experts.5.w2.codes": "model-00002-of-00003.safetensors",
1363
+ "model.layers.22.block_sparse_moe.experts.5.w2.scales": "model-00002-of-00003.safetensors",
1364
+ "model.layers.22.block_sparse_moe.experts.5.w3.codebooks": "model-00002-of-00003.safetensors",
1365
+ "model.layers.22.block_sparse_moe.experts.5.w3.codes": "model-00002-of-00003.safetensors",
1366
+ "model.layers.22.block_sparse_moe.experts.5.w3.scales": "model-00002-of-00003.safetensors",
1367
+ "model.layers.22.block_sparse_moe.experts.6.w1.codebooks": "model-00002-of-00003.safetensors",
1368
+ "model.layers.22.block_sparse_moe.experts.6.w1.codes": "model-00002-of-00003.safetensors",
1369
+ "model.layers.22.block_sparse_moe.experts.6.w1.scales": "model-00002-of-00003.safetensors",
1370
+ "model.layers.22.block_sparse_moe.experts.6.w2.codebooks": "model-00002-of-00003.safetensors",
1371
+ "model.layers.22.block_sparse_moe.experts.6.w2.codes": "model-00002-of-00003.safetensors",
1372
+ "model.layers.22.block_sparse_moe.experts.6.w2.scales": "model-00002-of-00003.safetensors",
1373
+ "model.layers.22.block_sparse_moe.experts.6.w3.codebooks": "model-00002-of-00003.safetensors",
1374
+ "model.layers.22.block_sparse_moe.experts.6.w3.codes": "model-00002-of-00003.safetensors",
1375
+ "model.layers.22.block_sparse_moe.experts.6.w3.scales": "model-00002-of-00003.safetensors",
1376
+ "model.layers.22.block_sparse_moe.experts.7.w1.codebooks": "model-00002-of-00003.safetensors",
1377
+ "model.layers.22.block_sparse_moe.experts.7.w1.codes": "model-00002-of-00003.safetensors",
1378
+ "model.layers.22.block_sparse_moe.experts.7.w1.scales": "model-00002-of-00003.safetensors",
1379
+ "model.layers.22.block_sparse_moe.experts.7.w2.codebooks": "model-00002-of-00003.safetensors",
1380
+ "model.layers.22.block_sparse_moe.experts.7.w2.codes": "model-00002-of-00003.safetensors",
1381
+ "model.layers.22.block_sparse_moe.experts.7.w2.scales": "model-00002-of-00003.safetensors",
1382
+ "model.layers.22.block_sparse_moe.experts.7.w3.codebooks": "model-00002-of-00003.safetensors",
1383
+ "model.layers.22.block_sparse_moe.experts.7.w3.codes": "model-00002-of-00003.safetensors",
1384
+ "model.layers.22.block_sparse_moe.experts.7.w3.scales": "model-00002-of-00003.safetensors",
1385
  "model.layers.22.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
1386
+ "model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
1387
+ "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
1388
  "model.layers.22.self_attn.k_proj.codebooks": "model-00002-of-00003.safetensors",
1389
  "model.layers.22.self_attn.k_proj.codes": "model-00002-of-00003.safetensors",
1390
  "model.layers.22.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
 
1397
  "model.layers.22.self_attn.v_proj.codebooks": "model-00002-of-00003.safetensors",
1398
  "model.layers.22.self_attn.v_proj.codes": "model-00002-of-00003.safetensors",
1399
  "model.layers.22.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
1400
+ "model.layers.23.block_sparse_moe.experts.0.w1.codebooks": "model-00002-of-00003.safetensors",
1401
+ "model.layers.23.block_sparse_moe.experts.0.w1.codes": "model-00002-of-00003.safetensors",
1402
+ "model.layers.23.block_sparse_moe.experts.0.w1.scales": "model-00002-of-00003.safetensors",
1403
+ "model.layers.23.block_sparse_moe.experts.0.w2.codebooks": "model-00002-of-00003.safetensors",
1404
+ "model.layers.23.block_sparse_moe.experts.0.w2.codes": "model-00002-of-00003.safetensors",
1405
+ "model.layers.23.block_sparse_moe.experts.0.w2.scales": "model-00002-of-00003.safetensors",
1406
+ "model.layers.23.block_sparse_moe.experts.0.w3.codebooks": "model-00002-of-00003.safetensors",
1407
+ "model.layers.23.block_sparse_moe.experts.0.w3.codes": "model-00002-of-00003.safetensors",
1408
+ "model.layers.23.block_sparse_moe.experts.0.w3.scales": "model-00002-of-00003.safetensors",
1409
+ "model.layers.23.block_sparse_moe.experts.1.w1.codebooks": "model-00002-of-00003.safetensors",
1410
+ "model.layers.23.block_sparse_moe.experts.1.w1.codes": "model-00002-of-00003.safetensors",
1411
+ "model.layers.23.block_sparse_moe.experts.1.w1.scales": "model-00002-of-00003.safetensors",
1412
+ "model.layers.23.block_sparse_moe.experts.1.w2.codebooks": "model-00002-of-00003.safetensors",
1413
+ "model.layers.23.block_sparse_moe.experts.1.w2.codes": "model-00002-of-00003.safetensors",
1414
+ "model.layers.23.block_sparse_moe.experts.1.w2.scales": "model-00002-of-00003.safetensors",
1415
+ "model.layers.23.block_sparse_moe.experts.1.w3.codebooks": "model-00002-of-00003.safetensors",
1416
+ "model.layers.23.block_sparse_moe.experts.1.w3.codes": "model-00002-of-00003.safetensors",
1417
+ "model.layers.23.block_sparse_moe.experts.1.w3.scales": "model-00002-of-00003.safetensors",
1418
+ "model.layers.23.block_sparse_moe.experts.2.w1.codebooks": "model-00002-of-00003.safetensors",
1419
+ "model.layers.23.block_sparse_moe.experts.2.w1.codes": "model-00002-of-00003.safetensors",
1420
+ "model.layers.23.block_sparse_moe.experts.2.w1.scales": "model-00002-of-00003.safetensors",
1421
+ "model.layers.23.block_sparse_moe.experts.2.w2.codebooks": "model-00002-of-00003.safetensors",
1422
+ "model.layers.23.block_sparse_moe.experts.2.w2.codes": "model-00002-of-00003.safetensors",
1423
+ "model.layers.23.block_sparse_moe.experts.2.w2.scales": "model-00002-of-00003.safetensors",
1424
+ "model.layers.23.block_sparse_moe.experts.2.w3.codebooks": "model-00002-of-00003.safetensors",
1425
+ "model.layers.23.block_sparse_moe.experts.2.w3.codes": "model-00002-of-00003.safetensors",
1426
+ "model.layers.23.block_sparse_moe.experts.2.w3.scales": "model-00002-of-00003.safetensors",
1427
+ "model.layers.23.block_sparse_moe.experts.3.w1.codebooks": "model-00002-of-00003.safetensors",
1428
+ "model.layers.23.block_sparse_moe.experts.3.w1.codes": "model-00002-of-00003.safetensors",
1429
+ "model.layers.23.block_sparse_moe.experts.3.w1.scales": "model-00002-of-00003.safetensors",
1430
+ "model.layers.23.block_sparse_moe.experts.3.w2.codebooks": "model-00002-of-00003.safetensors",
1431
+ "model.layers.23.block_sparse_moe.experts.3.w2.codes": "model-00002-of-00003.safetensors",
1432
+ "model.layers.23.block_sparse_moe.experts.3.w2.scales": "model-00002-of-00003.safetensors",
1433
+ "model.layers.23.block_sparse_moe.experts.3.w3.codebooks": "model-00002-of-00003.safetensors",
1434
+ "model.layers.23.block_sparse_moe.experts.3.w3.codes": "model-00002-of-00003.safetensors",
1435
+ "model.layers.23.block_sparse_moe.experts.3.w3.scales": "model-00002-of-00003.safetensors",
1436
+ "model.layers.23.block_sparse_moe.experts.4.w1.codebooks": "model-00002-of-00003.safetensors",
1437
+ "model.layers.23.block_sparse_moe.experts.4.w1.codes": "model-00002-of-00003.safetensors",
1438
+ "model.layers.23.block_sparse_moe.experts.4.w1.scales": "model-00002-of-00003.safetensors",
1439
+ "model.layers.23.block_sparse_moe.experts.4.w2.codebooks": "model-00002-of-00003.safetensors",
1440
+ "model.layers.23.block_sparse_moe.experts.4.w2.codes": "model-00002-of-00003.safetensors",
1441
+ "model.layers.23.block_sparse_moe.experts.4.w2.scales": "model-00002-of-00003.safetensors",
1442
+ "model.layers.23.block_sparse_moe.experts.4.w3.codebooks": "model-00002-of-00003.safetensors",
1443
+ "model.layers.23.block_sparse_moe.experts.4.w3.codes": "model-00002-of-00003.safetensors",
1444
+ "model.layers.23.block_sparse_moe.experts.4.w3.scales": "model-00002-of-00003.safetensors",
1445
+ "model.layers.23.block_sparse_moe.experts.5.w1.codebooks": "model-00002-of-00003.safetensors",
1446
+ "model.layers.23.block_sparse_moe.experts.5.w1.codes": "model-00002-of-00003.safetensors",
1447
+ "model.layers.23.block_sparse_moe.experts.5.w1.scales": "model-00002-of-00003.safetensors",
1448
+ "model.layers.23.block_sparse_moe.experts.5.w2.codebooks": "model-00002-of-00003.safetensors",
1449
+ "model.layers.23.block_sparse_moe.experts.5.w2.codes": "model-00002-of-00003.safetensors",
1450
+ "model.layers.23.block_sparse_moe.experts.5.w2.scales": "model-00002-of-00003.safetensors",
1451
+ "model.layers.23.block_sparse_moe.experts.5.w3.codebooks": "model-00002-of-00003.safetensors",
1452
+ "model.layers.23.block_sparse_moe.experts.5.w3.codes": "model-00002-of-00003.safetensors",
1453
+ "model.layers.23.block_sparse_moe.experts.5.w3.scales": "model-00002-of-00003.safetensors",
1454
+ "model.layers.23.block_sparse_moe.experts.6.w1.codebooks": "model-00002-of-00003.safetensors",
1455
+ "model.layers.23.block_sparse_moe.experts.6.w1.codes": "model-00002-of-00003.safetensors",
1456
+ "model.layers.23.block_sparse_moe.experts.6.w1.scales": "model-00002-of-00003.safetensors",
1457
+ "model.layers.23.block_sparse_moe.experts.6.w2.codebooks": "model-00002-of-00003.safetensors",
1458
+ "model.layers.23.block_sparse_moe.experts.6.w2.codes": "model-00002-of-00003.safetensors",
1459
+ "model.layers.23.block_sparse_moe.experts.6.w2.scales": "model-00002-of-00003.safetensors",
1460
+ "model.layers.23.block_sparse_moe.experts.6.w3.codebooks": "model-00002-of-00003.safetensors",
1461
+ "model.layers.23.block_sparse_moe.experts.6.w3.codes": "model-00002-of-00003.safetensors",
1462
+ "model.layers.23.block_sparse_moe.experts.6.w3.scales": "model-00002-of-00003.safetensors",
1463
+ "model.layers.23.block_sparse_moe.experts.7.w1.codebooks": "model-00002-of-00003.safetensors",
1464
+ "model.layers.23.block_sparse_moe.experts.7.w1.codes": "model-00002-of-00003.safetensors",
1465
+ "model.layers.23.block_sparse_moe.experts.7.w1.scales": "model-00002-of-00003.safetensors",
1466
+ "model.layers.23.block_sparse_moe.experts.7.w2.codebooks": "model-00002-of-00003.safetensors",
1467
+ "model.layers.23.block_sparse_moe.experts.7.w2.codes": "model-00002-of-00003.safetensors",
1468
+ "model.layers.23.block_sparse_moe.experts.7.w2.scales": "model-00002-of-00003.safetensors",
1469
+ "model.layers.23.block_sparse_moe.experts.7.w3.codebooks": "model-00002-of-00003.safetensors",
1470
+ "model.layers.23.block_sparse_moe.experts.7.w3.codes": "model-00002-of-00003.safetensors",
1471
+ "model.layers.23.block_sparse_moe.experts.7.w3.scales": "model-00002-of-00003.safetensors",
1472
+ "model.layers.23.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
1473
+ "model.layers.23.input_layernorm.weight": "model-00002-of-00003.safetensors",
1474
+ "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
1475
+ "model.layers.23.self_attn.k_proj.codebooks": "model-00002-of-00003.safetensors",
1476
+ "model.layers.23.self_attn.k_proj.codes": "model-00002-of-00003.safetensors",
1477
+ "model.layers.23.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
1478
+ "model.layers.23.self_attn.o_proj.codebooks": "model-00002-of-00003.safetensors",
1479
+ "model.layers.23.self_attn.o_proj.codes": "model-00002-of-00003.safetensors",
1480
+ "model.layers.23.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
1481
+ "model.layers.23.self_attn.q_proj.codebooks": "model-00002-of-00003.safetensors",
1482
+ "model.layers.23.self_attn.q_proj.codes": "model-00002-of-00003.safetensors",
1483
+ "model.layers.23.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
1484
+ "model.layers.23.self_attn.v_proj.codebooks": "model-00002-of-00003.safetensors",
1485
+ "model.layers.23.self_attn.v_proj.codes": "model-00002-of-00003.safetensors",
1486
+ "model.layers.23.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
1487
+ "model.layers.24.block_sparse_moe.experts.0.w1.codebooks": "model-00002-of-00003.safetensors",
1488
+ "model.layers.24.block_sparse_moe.experts.0.w1.codes": "model-00002-of-00003.safetensors",
1489
+ "model.layers.24.block_sparse_moe.experts.0.w1.scales": "model-00002-of-00003.safetensors",
1490
+ "model.layers.24.block_sparse_moe.experts.0.w2.codebooks": "model-00002-of-00003.safetensors",
1491
+ "model.layers.24.block_sparse_moe.experts.0.w2.codes": "model-00002-of-00003.safetensors",
1492
+ "model.layers.24.block_sparse_moe.experts.0.w2.scales": "model-00002-of-00003.safetensors",
1493
+ "model.layers.24.block_sparse_moe.experts.0.w3.codebooks": "model-00002-of-00003.safetensors",
1494
+ "model.layers.24.block_sparse_moe.experts.0.w3.codes": "model-00002-of-00003.safetensors",
1495
+ "model.layers.24.block_sparse_moe.experts.0.w3.scales": "model-00002-of-00003.safetensors",
1496
+ "model.layers.24.block_sparse_moe.experts.1.w1.codebooks": "model-00002-of-00003.safetensors",
1497
+ "model.layers.24.block_sparse_moe.experts.1.w1.codes": "model-00002-of-00003.safetensors",
1498
+ "model.layers.24.block_sparse_moe.experts.1.w1.scales": "model-00002-of-00003.safetensors",
1499
+ "model.layers.24.block_sparse_moe.experts.1.w2.codebooks": "model-00002-of-00003.safetensors",
1500
+ "model.layers.24.block_sparse_moe.experts.1.w2.codes": "model-00002-of-00003.safetensors",
1501
+ "model.layers.24.block_sparse_moe.experts.1.w2.scales": "model-00002-of-00003.safetensors",
1502
+ "model.layers.24.block_sparse_moe.experts.1.w3.codebooks": "model-00002-of-00003.safetensors",
1503
+ "model.layers.24.block_sparse_moe.experts.1.w3.codes": "model-00002-of-00003.safetensors",
1504
+ "model.layers.24.block_sparse_moe.experts.1.w3.scales": "model-00002-of-00003.safetensors",
1505
+ "model.layers.24.block_sparse_moe.experts.2.w1.codebooks": "model-00002-of-00003.safetensors",
1506
+ "model.layers.24.block_sparse_moe.experts.2.w1.codes": "model-00002-of-00003.safetensors",
1507
+ "model.layers.24.block_sparse_moe.experts.2.w1.scales": "model-00002-of-00003.safetensors",
1508
+ "model.layers.24.block_sparse_moe.experts.2.w2.codebooks": "model-00002-of-00003.safetensors",
1509
+ "model.layers.24.block_sparse_moe.experts.2.w2.codes": "model-00002-of-00003.safetensors",
1510
+ "model.layers.24.block_sparse_moe.experts.2.w2.scales": "model-00002-of-00003.safetensors",
1511
+ "model.layers.24.block_sparse_moe.experts.2.w3.codebooks": "model-00002-of-00003.safetensors",
1512
+ "model.layers.24.block_sparse_moe.experts.2.w3.codes": "model-00002-of-00003.safetensors",
1513
+ "model.layers.24.block_sparse_moe.experts.2.w3.scales": "model-00002-of-00003.safetensors",
1514
+ "model.layers.24.block_sparse_moe.experts.3.w1.codebooks": "model-00002-of-00003.safetensors",
1515
+ "model.layers.24.block_sparse_moe.experts.3.w1.codes": "model-00002-of-00003.safetensors",
1516
+ "model.layers.24.block_sparse_moe.experts.3.w1.scales": "model-00002-of-00003.safetensors",
1517
+ "model.layers.24.block_sparse_moe.experts.3.w2.codebooks": "model-00002-of-00003.safetensors",
1518
+ "model.layers.24.block_sparse_moe.experts.3.w2.codes": "model-00002-of-00003.safetensors",
1519
+ "model.layers.24.block_sparse_moe.experts.3.w2.scales": "model-00002-of-00003.safetensors",
1520
+ "model.layers.24.block_sparse_moe.experts.3.w3.codebooks": "model-00002-of-00003.safetensors",
1521
+ "model.layers.24.block_sparse_moe.experts.3.w3.codes": "model-00002-of-00003.safetensors",
1522
+ "model.layers.24.block_sparse_moe.experts.3.w3.scales": "model-00002-of-00003.safetensors",
1523
+ "model.layers.24.block_sparse_moe.experts.4.w1.codebooks": "model-00002-of-00003.safetensors",
1524
+ "model.layers.24.block_sparse_moe.experts.4.w1.codes": "model-00002-of-00003.safetensors",
1525
+ "model.layers.24.block_sparse_moe.experts.4.w1.scales": "model-00002-of-00003.safetensors",
1526
+ "model.layers.24.block_sparse_moe.experts.4.w2.codebooks": "model-00002-of-00003.safetensors",
1527
+ "model.layers.24.block_sparse_moe.experts.4.w2.codes": "model-00002-of-00003.safetensors",
1528
+ "model.layers.24.block_sparse_moe.experts.4.w2.scales": "model-00002-of-00003.safetensors",
1529
+ "model.layers.24.block_sparse_moe.experts.4.w3.codebooks": "model-00002-of-00003.safetensors",
1530
+ "model.layers.24.block_sparse_moe.experts.4.w3.codes": "model-00002-of-00003.safetensors",
1531
+ "model.layers.24.block_sparse_moe.experts.4.w3.scales": "model-00002-of-00003.safetensors",
1532
+ "model.layers.24.block_sparse_moe.experts.5.w1.codebooks": "model-00002-of-00003.safetensors",
1533
+ "model.layers.24.block_sparse_moe.experts.5.w1.codes": "model-00002-of-00003.safetensors",
1534
+ "model.layers.24.block_sparse_moe.experts.5.w1.scales": "model-00002-of-00003.safetensors",
1535
+ "model.layers.24.block_sparse_moe.experts.5.w2.codebooks": "model-00002-of-00003.safetensors",
1536
+ "model.layers.24.block_sparse_moe.experts.5.w2.codes": "model-00002-of-00003.safetensors",
1537
+ "model.layers.24.block_sparse_moe.experts.5.w2.scales": "model-00002-of-00003.safetensors",
1538
+ "model.layers.24.block_sparse_moe.experts.5.w3.codebooks": "model-00002-of-00003.safetensors",
1539
+ "model.layers.24.block_sparse_moe.experts.5.w3.codes": "model-00002-of-00003.safetensors",
1540
+ "model.layers.24.block_sparse_moe.experts.5.w3.scales": "model-00002-of-00003.safetensors",
1541
+ "model.layers.24.block_sparse_moe.experts.6.w1.codebooks": "model-00002-of-00003.safetensors",
1542
  "model.layers.24.block_sparse_moe.experts.6.w1.codes": "model-00003-of-00003.safetensors",
1543
  "model.layers.24.block_sparse_moe.experts.6.w1.scales": "model-00003-of-00003.safetensors",
1544
  "model.layers.24.block_sparse_moe.experts.6.w2.codebooks": "model-00003-of-00003.safetensors",
 
1556
  "model.layers.24.block_sparse_moe.experts.7.w3.codebooks": "model-00003-of-00003.safetensors",
1557
  "model.layers.24.block_sparse_moe.experts.7.w3.codes": "model-00003-of-00003.safetensors",
1558
  "model.layers.24.block_sparse_moe.experts.7.w3.scales": "model-00003-of-00003.safetensors",
1559
+ "model.layers.24.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
1560
  "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
1561
  "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
1562
+ "model.layers.24.self_attn.k_proj.codebooks": "model-00002-of-00003.safetensors",
1563
+ "model.layers.24.self_attn.k_proj.codes": "model-00002-of-00003.safetensors",
1564
+ "model.layers.24.self_attn.k_proj.scales": "model-00002-of-00003.safetensors",
1565
+ "model.layers.24.self_attn.o_proj.codebooks": "model-00002-of-00003.safetensors",
1566
+ "model.layers.24.self_attn.o_proj.codes": "model-00002-of-00003.safetensors",
1567
+ "model.layers.24.self_attn.o_proj.scales": "model-00002-of-00003.safetensors",
1568
+ "model.layers.24.self_attn.q_proj.codebooks": "model-00002-of-00003.safetensors",
1569
+ "model.layers.24.self_attn.q_proj.codes": "model-00002-of-00003.safetensors",
1570
+ "model.layers.24.self_attn.q_proj.scales": "model-00002-of-00003.safetensors",
1571
+ "model.layers.24.self_attn.v_proj.codebooks": "model-00002-of-00003.safetensors",
1572
+ "model.layers.24.self_attn.v_proj.codes": "model-00002-of-00003.safetensors",
1573
+ "model.layers.24.self_attn.v_proj.scales": "model-00002-of-00003.safetensors",
1574
  "model.layers.25.block_sparse_moe.experts.0.w1.codebooks": "model-00003-of-00003.safetensors",
1575
  "model.layers.25.block_sparse_moe.experts.0.w1.codes": "model-00003-of-00003.safetensors",
1576
  "model.layers.25.block_sparse_moe.experts.0.w1.scales": "model-00003-of-00003.safetensors",
special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "unk_token": "<unk>"
5
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [],
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": true,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": null,
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false,
42
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}"
43
+ }