Hugging Face's logo

Gen-Verse
/
MMaDA-8B-MixCoT

Any-to-Any
Transformers
Safetensors
llada
feature-extraction
custom_code
Model card Files Community
tyfeld commited on
Commit
8c9b50b
·
verified ·
1 Parent(s): 4954d83

Upload model

Browse files
Files changed (7) hide show
  1. config.json +64 -0
  2. generation_config.json +6 -0
  3. model-00001-of-00004.safetensors +3 -0
  4. model-00002-of-00004.safetensors +3 -0
  5. model-00003-of-00004.safetensors +3 -0
  6. model-00004-of-00004.safetensors +3 -0
  7. model.safetensors.index.json +298 -0
config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/data_storage/lbw/MMaDA/mmada-training-stage4-llada-instruct/checkpoint-270000/unwrapped_model",
3
+ "activation_type": "silu",
4
+ "alibi": false,
5
+ "alibi_bias_max": 8.0,
6
+ "architectures": [
7
+ "LLaDAModelLM"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "attention_layer_norm": false,
11
+ "attention_layer_norm_with_affine": true,
12
+ "auto_map": {
13
+ "AutoConfig": "configuration_llada.LLaDAConfig",
14
+ "AutoModel": "modeling_llada.LLaDAModelLM",
15
+ "AutoModelForCausalLM": "modeling_llada.LLaDAModelLM"
16
+ },
17
+ "bias_for_layer_norm": false,
18
+ "block_group_size": 1,
19
+ "block_type": "llama",
20
+ "codebook_size": 8192,
21
+ "d_model": 4096,
22
+ "embedding_dropout": 0.0,
23
+ "embedding_size": 134656,
24
+ "eos_token_id": 126081,
25
+ "flash_attention": false,
26
+ "include_bias": false,
27
+ "include_qkv_bias": false,
28
+ "init_cutoff_factor": null,
29
+ "init_device": "meta",
30
+ "init_fn": "mitchell",
31
+ "init_std": 0.02,
32
+ "input_emb_norm": false,
33
+ "layer_norm_type": "rms",
34
+ "layer_norm_with_affine": true,
35
+ "llm_vocab_size": 126464,
36
+ "mask_token_id": 126336,
37
+ "max_sequence_length": 4096,
38
+ "mlp_hidden_size": 12288,
39
+ "mlp_ratio": 4,
40
+ "model_type": "llada",
41
+ "multi_query_attention": null,
42
+ "n_heads": 32,
43
+ "n_kv_heads": 32,
44
+ "n_layers": 32,
45
+ "new_vocab_size": 134656,
46
+ "num_new_special_tokens": 0,
47
+ "num_vq_tokens": 256,
48
+ "pad_token_id": 126081,
49
+ "precision": "amp_bf16",
50
+ "pretrained_model_path": "/data_storage/shared/pretrained_models/LLaDA-8B-Instruct",
51
+ "residual_dropout": 0.0,
52
+ "rms_norm_eps": 1e-05,
53
+ "rope": true,
54
+ "rope_full_precision": true,
55
+ "rope_theta": 500000.0,
56
+ "scale_logits": false,
57
+ "tie_word_embeddings": false,
58
+ "torch_dtype": "bfloat16",
59
+ "transformers_version": "4.46.0",
60
+ "use_cache": false,
61
+ "vocab_size": 134656,
62
+ "w_clip_vit": false,
63
+ "weight_tying": false
64
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 126080,
4
+ "eos_token_id": 126081,
5
+ "transformers_version": "4.46.0"
6
+ }
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83cfbf5001399c50b0e6c3ce8ce9cf7e11296c203563845a3527bfb5e7b65d78
3
+ size 4928472568
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21a0fc29533d4db67a511b2edba82e9ba118e240152296ee5aa9e142184873cd
3
+ size 4932693832
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b35559803b4eb2facd73749237a4472da20bca7a2607d6b5a4b4a416e2370aab
3
+ size 4999819560
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc78561e26f88350f6c6b405c8487b297c5c5f5a44533b53a09c8450c33bad8f
3
+ size 1304428928
model.safetensors.index.json ADDED
@@ -0,0 +1,298 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 16165380096
4
+ },
5
+ "weight_map": {
6
+ "model.transformer.blocks.0.attn_norm.weight": "model-00001-of-00004.safetensors",
7
+ "model.transformer.blocks.0.attn_out.weight": "model-00001-of-00004.safetensors",
8
+ "model.transformer.blocks.0.ff_norm.weight": "model-00001-of-00004.safetensors",
9
+ "model.transformer.blocks.0.ff_out.weight": "model-00001-of-00004.safetensors",
10
+ "model.transformer.blocks.0.ff_proj.weight": "model-00001-of-00004.safetensors",
11
+ "model.transformer.blocks.0.k_proj.weight": "model-00001-of-00004.safetensors",
12
+ "model.transformer.blocks.0.q_proj.weight": "model-00001-of-00004.safetensors",
13
+ "model.transformer.blocks.0.up_proj.weight": "model-00001-of-00004.safetensors",
14
+ "model.transformer.blocks.0.v_proj.weight": "model-00001-of-00004.safetensors",
15
+ "model.transformer.blocks.1.attn_norm.weight": "model-00001-of-00004.safetensors",
16
+ "model.transformer.blocks.1.attn_out.weight": "model-00001-of-00004.safetensors",
17
+ "model.transformer.blocks.1.ff_norm.weight": "model-00001-of-00004.safetensors",
18
+ "model.transformer.blocks.1.ff_out.weight": "model-00001-of-00004.safetensors",
19
+ "model.transformer.blocks.1.ff_proj.weight": "model-00001-of-00004.safetensors",
20
+ "model.transformer.blocks.1.k_proj.weight": "model-00001-of-00004.safetensors",
21
+ "model.transformer.blocks.1.q_proj.weight": "model-00001-of-00004.safetensors",
22
+ "model.transformer.blocks.1.up_proj.weight": "model-00001-of-00004.safetensors",
23
+ "model.transformer.blocks.1.v_proj.weight": "model-00001-of-00004.safetensors",
24
+ "model.transformer.blocks.10.attn_norm.weight": "model-00002-of-00004.safetensors",
25
+ "model.transformer.blocks.10.attn_out.weight": "model-00002-of-00004.safetensors",
26
+ "model.transformer.blocks.10.ff_norm.weight": "model-00002-of-00004.safetensors",
27
+ "model.transformer.blocks.10.ff_out.weight": "model-00002-of-00004.safetensors",
28
+ "model.transformer.blocks.10.ff_proj.weight": "model-00002-of-00004.safetensors",
29
+ "model.transformer.blocks.10.k_proj.weight": "model-00002-of-00004.safetensors",
30
+ "model.transformer.blocks.10.q_proj.weight": "model-00002-of-00004.safetensors",
31
+ "model.transformer.blocks.10.up_proj.weight": "model-00002-of-00004.safetensors",
32
+ "model.transformer.blocks.10.v_proj.weight": "model-00002-of-00004.safetensors",
33
+ "model.transformer.blocks.11.attn_norm.weight": "model-00002-of-00004.safetensors",
34
+ "model.transformer.blocks.11.attn_out.weight": "model-00002-of-00004.safetensors",
35
+ "model.transformer.blocks.11.ff_norm.weight": "model-00002-of-00004.safetensors",
36
+ "model.transformer.blocks.11.ff_out.weight": "model-00002-of-00004.safetensors",
37
+ "model.transformer.blocks.11.ff_proj.weight": "model-00002-of-00004.safetensors",
38
+ "model.transformer.blocks.11.k_proj.weight": "model-00002-of-00004.safetensors",
39
+ "model.transformer.blocks.11.q_proj.weight": "model-00002-of-00004.safetensors",
40
+ "model.transformer.blocks.11.up_proj.weight": "model-00002-of-00004.safetensors",
41
+ "model.transformer.blocks.11.v_proj.weight": "model-00002-of-00004.safetensors",
42
+ "model.transformer.blocks.12.attn_norm.weight": "model-00002-of-00004.safetensors",
43
+ "model.transformer.blocks.12.attn_out.weight": "model-00002-of-00004.safetensors",
44
+ "model.transformer.blocks.12.ff_norm.weight": "model-00002-of-00004.safetensors",
45
+ "model.transformer.blocks.12.ff_out.weight": "model-00002-of-00004.safetensors",
46
+ "model.transformer.blocks.12.ff_proj.weight": "model-00002-of-00004.safetensors",
47
+ "model.transformer.blocks.12.k_proj.weight": "model-00002-of-00004.safetensors",
48
+ "model.transformer.blocks.12.q_proj.weight": "model-00002-of-00004.safetensors",
49
+ "model.transformer.blocks.12.up_proj.weight": "model-00002-of-00004.safetensors",
50
+ "model.transformer.blocks.12.v_proj.weight": "model-00002-of-00004.safetensors",
51
+ "model.transformer.blocks.13.attn_norm.weight": "model-00002-of-00004.safetensors",
52
+ "model.transformer.blocks.13.attn_out.weight": "model-00002-of-00004.safetensors",
53
+ "model.transformer.blocks.13.ff_norm.weight": "model-00002-of-00004.safetensors",
54
+ "model.transformer.blocks.13.ff_out.weight": "model-00002-of-00004.safetensors",
55
+ "model.transformer.blocks.13.ff_proj.weight": "model-00002-of-00004.safetensors",
56
+ "model.transformer.blocks.13.k_proj.weight": "model-00002-of-00004.safetensors",
57
+ "model.transformer.blocks.13.q_proj.weight": "model-00002-of-00004.safetensors",
58
+ "model.transformer.blocks.13.up_proj.weight": "model-00002-of-00004.safetensors",
59
+ "model.transformer.blocks.13.v_proj.weight": "model-00002-of-00004.safetensors",
60
+ "model.transformer.blocks.14.attn_norm.weight": "model-00002-of-00004.safetensors",
61
+ "model.transformer.blocks.14.attn_out.weight": "model-00002-of-00004.safetensors",
62
+ "model.transformer.blocks.14.ff_norm.weight": "model-00002-of-00004.safetensors",
63
+ "model.transformer.blocks.14.ff_out.weight": "model-00002-of-00004.safetensors",
64
+ "model.transformer.blocks.14.ff_proj.weight": "model-00002-of-00004.safetensors",
65
+ "model.transformer.blocks.14.k_proj.weight": "model-00002-of-00004.safetensors",
66
+ "model.transformer.blocks.14.q_proj.weight": "model-00002-of-00004.safetensors",
67
+ "model.transformer.blocks.14.up_proj.weight": "model-00002-of-00004.safetensors",
68
+ "model.transformer.blocks.14.v_proj.weight": "model-00002-of-00004.safetensors",
69
+ "model.transformer.blocks.15.attn_norm.weight": "model-00002-of-00004.safetensors",
70
+ "model.transformer.blocks.15.attn_out.weight": "model-00002-of-00004.safetensors",
71
+ "model.transformer.blocks.15.ff_norm.weight": "model-00002-of-00004.safetensors",
72
+ "model.transformer.blocks.15.ff_out.weight": "model-00002-of-00004.safetensors",
73
+ "model.transformer.blocks.15.ff_proj.weight": "model-00002-of-00004.safetensors",
74
+ "model.transformer.blocks.15.k_proj.weight": "model-00002-of-00004.safetensors",
75
+ "model.transformer.blocks.15.q_proj.weight": "model-00002-of-00004.safetensors",
76
+ "model.transformer.blocks.15.up_proj.weight": "model-00002-of-00004.safetensors",
77
+ "model.transformer.blocks.15.v_proj.weight": "model-00002-of-00004.safetensors",
78
+ "model.transformer.blocks.16.attn_norm.weight": "model-00002-of-00004.safetensors",
79
+ "model.transformer.blocks.16.attn_out.weight": "model-00002-of-00004.safetensors",
80
+ "model.transformer.blocks.16.ff_norm.weight": "model-00002-of-00004.safetensors",
81
+ "model.transformer.blocks.16.ff_out.weight": "model-00002-of-00004.safetensors",
82
+ "model.transformer.blocks.16.ff_proj.weight": "model-00002-of-00004.safetensors",
83
+ "model.transformer.blocks.16.k_proj.weight": "model-00002-of-00004.safetensors",
84
+ "model.transformer.blocks.16.q_proj.weight": "model-00002-of-00004.safetensors",
85
+ "model.transformer.blocks.16.up_proj.weight": "model-00002-of-00004.safetensors",
86
+ "model.transformer.blocks.16.v_proj.weight": "model-00002-of-00004.safetensors",
87
+ "model.transformer.blocks.17.attn_norm.weight": "model-00002-of-00004.safetensors",
88
+ "model.transformer.blocks.17.attn_out.weight": "model-00002-of-00004.safetensors",
89
+ "model.transformer.blocks.17.ff_norm.weight": "model-00002-of-00004.safetensors",
90
+ "model.transformer.blocks.17.ff_out.weight": "model-00002-of-00004.safetensors",
91
+ "model.transformer.blocks.17.ff_proj.weight": "model-00002-of-00004.safetensors",
92
+ "model.transformer.blocks.17.k_proj.weight": "model-00002-of-00004.safetensors",
93
+ "model.transformer.blocks.17.q_proj.weight": "model-00002-of-00004.safetensors",
94
+ "model.transformer.blocks.17.up_proj.weight": "model-00002-of-00004.safetensors",
95
+ "model.transformer.blocks.17.v_proj.weight": "model-00002-of-00004.safetensors",
96
+ "model.transformer.blocks.18.attn_norm.weight": "model-00002-of-00004.safetensors",
97
+ "model.transformer.blocks.18.attn_out.weight": "model-00002-of-00004.safetensors",
98
+ "model.transformer.blocks.18.ff_norm.weight": "model-00002-of-00004.safetensors",
99
+ "model.transformer.blocks.18.ff_out.weight": "model-00002-of-00004.safetensors",
100
+ "model.transformer.blocks.18.ff_proj.weight": "model-00002-of-00004.safetensors",
101
+ "model.transformer.blocks.18.k_proj.weight": "model-00002-of-00004.safetensors",
102
+ "model.transformer.blocks.18.q_proj.weight": "model-00002-of-00004.safetensors",
103
+ "model.transformer.blocks.18.up_proj.weight": "model-00002-of-00004.safetensors",
104
+ "model.transformer.blocks.18.v_proj.weight": "model-00002-of-00004.safetensors",
105
+ "model.transformer.blocks.19.attn_norm.weight": "model-00002-of-00004.safetensors",
106
+ "model.transformer.blocks.19.attn_out.weight": "model-00002-of-00004.safetensors",
107
+ "model.transformer.blocks.19.ff_norm.weight": "model-00002-of-00004.safetensors",
108
+ "model.transformer.blocks.19.ff_out.weight": "model-00002-of-00004.safetensors",
109
+ "model.transformer.blocks.19.ff_proj.weight": "model-00002-of-00004.safetensors",
110
+ "model.transformer.blocks.19.k_proj.weight": "model-00002-of-00004.safetensors",
111
+ "model.transformer.blocks.19.q_proj.weight": "model-00002-of-00004.safetensors",
112
+ "model.transformer.blocks.19.up_proj.weight": "model-00002-of-00004.safetensors",
113
+ "model.transformer.blocks.19.v_proj.weight": "model-00002-of-00004.safetensors",
114
+ "model.transformer.blocks.2.attn_norm.weight": "model-00001-of-00004.safetensors",
115
+ "model.transformer.blocks.2.attn_out.weight": "model-00001-of-00004.safetensors",
116
+ "model.transformer.blocks.2.ff_norm.weight": "model-00001-of-00004.safetensors",
117
+ "model.transformer.blocks.2.ff_out.weight": "model-00001-of-00004.safetensors",
118
+ "model.transformer.blocks.2.ff_proj.weight": "model-00001-of-00004.safetensors",
119
+ "model.transformer.blocks.2.k_proj.weight": "model-00001-of-00004.safetensors",
120
+ "model.transformer.blocks.2.q_proj.weight": "model-00001-of-00004.safetensors",
121
+ "model.transformer.blocks.2.up_proj.weight": "model-00001-of-00004.safetensors",
122
+ "model.transformer.blocks.2.v_proj.weight": "model-00001-of-00004.safetensors",
123
+ "model.transformer.blocks.20.attn_norm.weight": "model-00003-of-00004.safetensors",
124
+ "model.transformer.blocks.20.attn_out.weight": "model-00002-of-00004.safetensors",
125
+ "model.transformer.blocks.20.ff_norm.weight": "model-00003-of-00004.safetensors",
126
+ "model.transformer.blocks.20.ff_out.weight": "model-00003-of-00004.safetensors",
127
+ "model.transformer.blocks.20.ff_proj.weight": "model-00003-of-00004.safetensors",
128
+ "model.transformer.blocks.20.k_proj.weight": "model-00003-of-00004.safetensors",
129
+ "model.transformer.blocks.20.q_proj.weight": "model-00003-of-00004.safetensors",
130
+ "model.transformer.blocks.20.up_proj.weight": "model-00003-of-00004.safetensors",
131
+ "model.transformer.blocks.20.v_proj.weight": "model-00003-of-00004.safetensors",
132
+ "model.transformer.blocks.21.attn_norm.weight": "model-00003-of-00004.safetensors",
133
+ "model.transformer.blocks.21.attn_out.weight": "model-00003-of-00004.safetensors",
134
+ "model.transformer.blocks.21.ff_norm.weight": "model-00003-of-00004.safetensors",
135
+ "model.transformer.blocks.21.ff_out.weight": "model-00003-of-00004.safetensors",
136
+ "model.transformer.blocks.21.ff_proj.weight": "model-00003-of-00004.safetensors",
137
+ "model.transformer.blocks.21.k_proj.weight": "model-00003-of-00004.safetensors",
138
+ "model.transformer.blocks.21.q_proj.weight": "model-00003-of-00004.safetensors",
139
+ "model.transformer.blocks.21.up_proj.weight": "model-00003-of-00004.safetensors",
140
+ "model.transformer.blocks.21.v_proj.weight": "model-00003-of-00004.safetensors",
141
+ "model.transformer.blocks.22.attn_norm.weight": "model-00003-of-00004.safetensors",
142
+ "model.transformer.blocks.22.attn_out.weight": "model-00003-of-00004.safetensors",
143
+ "model.transformer.blocks.22.ff_norm.weight": "model-00003-of-00004.safetensors",
144
+ "model.transformer.blocks.22.ff_out.weight": "model-00003-of-00004.safetensors",
145
+ "model.transformer.blocks.22.ff_proj.weight": "model-00003-of-00004.safetensors",
146
+ "model.transformer.blocks.22.k_proj.weight": "model-00003-of-00004.safetensors",
147
+ "model.transformer.blocks.22.q_proj.weight": "model-00003-of-00004.safetensors",
148
+ "model.transformer.blocks.22.up_proj.weight": "model-00003-of-00004.safetensors",
149
+ "model.transformer.blocks.22.v_proj.weight": "model-00003-of-00004.safetensors",
150
+ "model.transformer.blocks.23.attn_norm.weight": "model-00003-of-00004.safetensors",
151
+ "model.transformer.blocks.23.attn_out.weight": "model-00003-of-00004.safetensors",
152
+ "model.transformer.blocks.23.ff_norm.weight": "model-00003-of-00004.safetensors",
153
+ "model.transformer.blocks.23.ff_out.weight": "model-00003-of-00004.safetensors",
154
+ "model.transformer.blocks.23.ff_proj.weight": "model-00003-of-00004.safetensors",
155
+ "model.transformer.blocks.23.k_proj.weight": "model-00003-of-00004.safetensors",
156
+ "model.transformer.blocks.23.q_proj.weight": "model-00003-of-00004.safetensors",
157
+ "model.transformer.blocks.23.up_proj.weight": "model-00003-of-00004.safetensors",
158
+ "model.transformer.blocks.23.v_proj.weight": "model-00003-of-00004.safetensors",
159
+ "model.transformer.blocks.24.attn_norm.weight": "model-00003-of-00004.safetensors",
160
+ "model.transformer.blocks.24.attn_out.weight": "model-00003-of-00004.safetensors",
161
+ "model.transformer.blocks.24.ff_norm.weight": "model-00003-of-00004.safetensors",
162
+ "model.transformer.blocks.24.ff_out.weight": "model-00003-of-00004.safetensors",
163
+ "model.transformer.blocks.24.ff_proj.weight": "model-00003-of-00004.safetensors",
164
+ "model.transformer.blocks.24.k_proj.weight": "model-00003-of-00004.safetensors",
165
+ "model.transformer.blocks.24.q_proj.weight": "model-00003-of-00004.safetensors",
166
+ "model.transformer.blocks.24.up_proj.weight": "model-00003-of-00004.safetensors",
167
+ "model.transformer.blocks.24.v_proj.weight": "model-00003-of-00004.safetensors",
168
+ "model.transformer.blocks.25.attn_norm.weight": "model-00003-of-00004.safetensors",
169
+ "model.transformer.blocks.25.attn_out.weight": "model-00003-of-00004.safetensors",
170
+ "model.transformer.blocks.25.ff_norm.weight": "model-00003-of-00004.safetensors",
171
+ "model.transformer.blocks.25.ff_out.weight": "model-00003-of-00004.safetensors",
172
+ "model.transformer.blocks.25.ff_proj.weight": "model-00003-of-00004.safetensors",
173
+ "model.transformer.blocks.25.k_proj.weight": "model-00003-of-00004.safetensors",
174
+ "model.transformer.blocks.25.q_proj.weight": "model-00003-of-00004.safetensors",
175
+ "model.transformer.blocks.25.up_proj.weight": "model-00003-of-00004.safetensors",
176
+ "model.transformer.blocks.25.v_proj.weight": "model-00003-of-00004.safetensors",
177
+ "model.transformer.blocks.26.attn_norm.weight": "model-00003-of-00004.safetensors",
178
+ "model.transformer.blocks.26.attn_out.weight": "model-00003-of-00004.safetensors",
179
+ "model.transformer.blocks.26.ff_norm.weight": "model-00003-of-00004.safetensors",
180
+ "model.transformer.blocks.26.ff_out.weight": "model-00003-of-00004.safetensors",
181
+ "model.transformer.blocks.26.ff_proj.weight": "model-00003-of-00004.safetensors",
182
+ "model.transformer.blocks.26.k_proj.weight": "model-00003-of-00004.safetensors",
183
+ "model.transformer.blocks.26.q_proj.weight": "model-00003-of-00004.safetensors",
184
+ "model.transformer.blocks.26.up_proj.weight": "model-00003-of-00004.safetensors",
185
+ "model.transformer.blocks.26.v_proj.weight": "model-00003-of-00004.safetensors",
186
+ "model.transformer.blocks.27.attn_norm.weight": "model-00003-of-00004.safetensors",
187
+ "model.transformer.blocks.27.attn_out.weight": "model-00003-of-00004.safetensors",
188
+ "model.transformer.blocks.27.ff_norm.weight": "model-00003-of-00004.safetensors",
189
+ "model.transformer.blocks.27.ff_out.weight": "model-00003-of-00004.safetensors",
190
+ "model.transformer.blocks.27.ff_proj.weight": "model-00003-of-00004.safetensors",
191
+ "model.transformer.blocks.27.k_proj.weight": "model-00003-of-00004.safetensors",
192
+ "model.transformer.blocks.27.q_proj.weight": "model-00003-of-00004.safetensors",
193
+ "model.transformer.blocks.27.up_proj.weight": "model-00003-of-00004.safetensors",
194
+ "model.transformer.blocks.27.v_proj.weight": "model-00003-of-00004.safetensors",
195
+ "model.transformer.blocks.28.attn_norm.weight": "model-00003-of-00004.safetensors",
196
+ "model.transformer.blocks.28.attn_out.weight": "model-00003-of-00004.safetensors",
197
+ "model.transformer.blocks.28.ff_norm.weight": "model-00003-of-00004.safetensors",
198
+ "model.transformer.blocks.28.ff_out.weight": "model-00003-of-00004.safetensors",
199
+ "model.transformer.blocks.28.ff_proj.weight": "model-00003-of-00004.safetensors",
200
+ "model.transformer.blocks.28.k_proj.weight": "model-00003-of-00004.safetensors",
201
+ "model.transformer.blocks.28.q_proj.weight": "model-00003-of-00004.safetensors",
202
+ "model.transformer.blocks.28.up_proj.weight": "model-00003-of-00004.safetensors",
203
+ "model.transformer.blocks.28.v_proj.weight": "model-00003-of-00004.safetensors",
204
+ "model.transformer.blocks.29.attn_norm.weight": "model-00003-of-00004.safetensors",
205
+ "model.transformer.blocks.29.attn_out.weight": "model-00003-of-00004.safetensors",
206
+ "model.transformer.blocks.29.ff_norm.weight": "model-00003-of-00004.safetensors",
207
+ "model.transformer.blocks.29.ff_out.weight": "model-00003-of-00004.safetensors",
208
+ "model.transformer.blocks.29.ff_proj.weight": "model-00003-of-00004.safetensors",
209
+ "model.transformer.blocks.29.k_proj.weight": "model-00003-of-00004.safetensors",
210
+ "model.transformer.blocks.29.q_proj.weight": "model-00003-of-00004.safetensors",
211
+ "model.transformer.blocks.29.up_proj.weight": "model-00003-of-00004.safetensors",
212
+ "model.transformer.blocks.29.v_proj.weight": "model-00003-of-00004.safetensors",
213
+ "model.transformer.blocks.3.attn_norm.weight": "model-00001-of-00004.safetensors",
214
+ "model.transformer.blocks.3.attn_out.weight": "model-00001-of-00004.safetensors",
215
+ "model.transformer.blocks.3.ff_norm.weight": "model-00001-of-00004.safetensors",
216
+ "model.transformer.blocks.3.ff_out.weight": "model-00001-of-00004.safetensors",
217
+ "model.transformer.blocks.3.ff_proj.weight": "model-00001-of-00004.safetensors",
218
+ "model.transformer.blocks.3.k_proj.weight": "model-00001-of-00004.safetensors",
219
+ "model.transformer.blocks.3.q_proj.weight": "model-00001-of-00004.safetensors",
220
+ "model.transformer.blocks.3.up_proj.weight": "model-00001-of-00004.safetensors",
221
+ "model.transformer.blocks.3.v_proj.weight": "model-00001-of-00004.safetensors",
222
+ "model.transformer.blocks.30.attn_norm.weight": "model-00003-of-00004.safetensors",
223
+ "model.transformer.blocks.30.attn_out.weight": "model-00003-of-00004.safetensors",
224
+ "model.transformer.blocks.30.ff_norm.weight": "model-00003-of-00004.safetensors",
225
+ "model.transformer.blocks.30.ff_out.weight": "model-00003-of-00004.safetensors",
226
+ "model.transformer.blocks.30.ff_proj.weight": "model-00003-of-00004.safetensors",
227
+ "model.transformer.blocks.30.k_proj.weight": "model-00003-of-00004.safetensors",
228
+ "model.transformer.blocks.30.q_proj.weight": "model-00003-of-00004.safetensors",
229
+ "model.transformer.blocks.30.up_proj.weight": "model-00003-of-00004.safetensors",
230
+ "model.transformer.blocks.30.v_proj.weight": "model-00003-of-00004.safetensors",
231
+ "model.transformer.blocks.31.attn_norm.weight": "model-00003-of-00004.safetensors",
232
+ "model.transformer.blocks.31.attn_out.weight": "model-00003-of-00004.safetensors",
233
+ "model.transformer.blocks.31.ff_norm.weight": "model-00003-of-00004.safetensors",
234
+ "model.transformer.blocks.31.ff_out.weight": "model-00003-of-00004.safetensors",
235
+ "model.transformer.blocks.31.ff_proj.weight": "model-00004-of-00004.safetensors",
236
+ "model.transformer.blocks.31.k_proj.weight": "model-00003-of-00004.safetensors",
237
+ "model.transformer.blocks.31.q_proj.weight": "model-00003-of-00004.safetensors",
238
+ "model.transformer.blocks.31.up_proj.weight": "model-00004-of-00004.safetensors",
239
+ "model.transformer.blocks.31.v_proj.weight": "model-00003-of-00004.safetensors",
240
+ "model.transformer.blocks.4.attn_norm.weight": "model-00001-of-00004.safetensors",
241
+ "model.transformer.blocks.4.attn_out.weight": "model-00001-of-00004.safetensors",
242
+ "model.transformer.blocks.4.ff_norm.weight": "model-00001-of-00004.safetensors",
243
+ "model.transformer.blocks.4.ff_out.weight": "model-00001-of-00004.safetensors",
244
+ "model.transformer.blocks.4.ff_proj.weight": "model-00001-of-00004.safetensors",
245
+ "model.transformer.blocks.4.k_proj.weight": "model-00001-of-00004.safetensors",
246
+ "model.transformer.blocks.4.q_proj.weight": "model-00001-of-00004.safetensors",
247
+ "model.transformer.blocks.4.up_proj.weight": "model-00001-of-00004.safetensors",
248
+ "model.transformer.blocks.4.v_proj.weight": "model-00001-of-00004.safetensors",
249
+ "model.transformer.blocks.5.attn_norm.weight": "model-00001-of-00004.safetensors",
250
+ "model.transformer.blocks.5.attn_out.weight": "model-00001-of-00004.safetensors",
251
+ "model.transformer.blocks.5.ff_norm.weight": "model-00001-of-00004.safetensors",
252
+ "model.transformer.blocks.5.ff_out.weight": "model-00001-of-00004.safetensors",
253
+ "model.transformer.blocks.5.ff_proj.weight": "model-00001-of-00004.safetensors",
254
+ "model.transformer.blocks.5.k_proj.weight": "model-00001-of-00004.safetensors",
255
+ "model.transformer.blocks.5.q_proj.weight": "model-00001-of-00004.safetensors",
256
+ "model.transformer.blocks.5.up_proj.weight": "model-00001-of-00004.safetensors",
257
+ "model.transformer.blocks.5.v_proj.weight": "model-00001-of-00004.safetensors",
258
+ "model.transformer.blocks.6.attn_norm.weight": "model-00001-of-00004.safetensors",
259
+ "model.transformer.blocks.6.attn_out.weight": "model-00001-of-00004.safetensors",
260
+ "model.transformer.blocks.6.ff_norm.weight": "model-00001-of-00004.safetensors",
261
+ "model.transformer.blocks.6.ff_out.weight": "model-00001-of-00004.safetensors",
262
+ "model.transformer.blocks.6.ff_proj.weight": "model-00001-of-00004.safetensors",
263
+ "model.transformer.blocks.6.k_proj.weight": "model-00001-of-00004.safetensors",
264
+ "model.transformer.blocks.6.q_proj.weight": "model-00001-of-00004.safetensors",
265
+ "model.transformer.blocks.6.up_proj.weight": "model-00001-of-00004.safetensors",
266
+ "model.transformer.blocks.6.v_proj.weight": "model-00001-of-00004.safetensors",
267
+ "model.transformer.blocks.7.attn_norm.weight": "model-00001-of-00004.safetensors",
268
+ "model.transformer.blocks.7.attn_out.weight": "model-00001-of-00004.safetensors",
269
+ "model.transformer.blocks.7.ff_norm.weight": "model-00001-of-00004.safetensors",
270
+ "model.transformer.blocks.7.ff_out.weight": "model-00001-of-00004.safetensors",
271
+ "model.transformer.blocks.7.ff_proj.weight": "model-00001-of-00004.safetensors",
272
+ "model.transformer.blocks.7.k_proj.weight": "model-00001-of-00004.safetensors",
273
+ "model.transformer.blocks.7.q_proj.weight": "model-00001-of-00004.safetensors",
274
+ "model.transformer.blocks.7.up_proj.weight": "model-00001-of-00004.safetensors",
275
+ "model.transformer.blocks.7.v_proj.weight": "model-00001-of-00004.safetensors",
276
+ "model.transformer.blocks.8.attn_norm.weight": "model-00001-of-00004.safetensors",
277
+ "model.transformer.blocks.8.attn_out.weight": "model-00001-of-00004.safetensors",
278
+ "model.transformer.blocks.8.ff_norm.weight": "model-00001-of-00004.safetensors",
279
+ "model.transformer.blocks.8.ff_out.weight": "model-00001-of-00004.safetensors",
280
+ "model.transformer.blocks.8.ff_proj.weight": "model-00001-of-00004.safetensors",
281
+ "model.transformer.blocks.8.k_proj.weight": "model-00001-of-00004.safetensors",
282
+ "model.transformer.blocks.8.q_proj.weight": "model-00001-of-00004.safetensors",
283
+ "model.transformer.blocks.8.up_proj.weight": "model-00002-of-00004.safetensors",
284
+ "model.transformer.blocks.8.v_proj.weight": "model-00001-of-00004.safetensors",
285
+ "model.transformer.blocks.9.attn_norm.weight": "model-00002-of-00004.safetensors",
286
+ "model.transformer.blocks.9.attn_out.weight": "model-00002-of-00004.safetensors",
287
+ "model.transformer.blocks.9.ff_norm.weight": "model-00002-of-00004.safetensors",
288
+ "model.transformer.blocks.9.ff_out.weight": "model-00002-of-00004.safetensors",
289
+ "model.transformer.blocks.9.ff_proj.weight": "model-00002-of-00004.safetensors",
290
+ "model.transformer.blocks.9.k_proj.weight": "model-00002-of-00004.safetensors",
291
+ "model.transformer.blocks.9.q_proj.weight": "model-00002-of-00004.safetensors",
292
+ "model.transformer.blocks.9.up_proj.weight": "model-00002-of-00004.safetensors",
293
+ "model.transformer.blocks.9.v_proj.weight": "model-00002-of-00004.safetensors",
294
+ "model.transformer.ff_out.weight": "model-00004-of-00004.safetensors",
295
+ "model.transformer.ln_f.weight": "model-00001-of-00004.safetensors",
296
+ "model.transformer.wte.weight": "model-00001-of-00004.safetensors"
297
+ }
298
+ }