juliusadebayo commited on
Commit
f1ff646
·
verified ·
1 Parent(s): bacea0d

Upload Steerling-8B weights (bf16)

Browse files
config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_type": "causal_diffusion",
3
+ "interpretable": true,
4
+ "n_layers": 32,
5
+ "n_head": 32,
6
+ "n_embd": 4096,
7
+ "block_size": 4096,
8
+ "n_kv_heads": 4,
9
+ "diff_block_size": 64,
10
+ "use_rms_norm": true,
11
+ "norm_eps": 1e-05,
12
+ "norm_order": "post",
13
+ "use_qk_norm": true,
14
+ "use_rope": true,
15
+ "rope_base": 500000.0,
16
+ "rope_full_precision": true,
17
+ "mlp_type": "swiglu",
18
+ "activation": "gelu",
19
+ "mlp_ratio": 4,
20
+ "intermediate_size": null,
21
+ "use_bias": false,
22
+ "clip_qkv": 10.0,
23
+ "weight_sharing": true,
24
+ "pad_token_id": 100277,
25
+ "bos_token_id": 100278,
26
+ "eos_token_id": 100257,
27
+ "endofchunk_token_id": 100279,
28
+ "mask_token_id": 100280,
29
+ "vocab_size": 100281,
30
+ "concept": {
31
+ "n_concepts": 33732,
32
+ "n_unknown_concepts": 101196,
33
+ "max_concepts": 16,
34
+ "concept_dim": 4096,
35
+ "use_attention_known": false,
36
+ "use_attention_unknown": false,
37
+ "topk_known": 16,
38
+ "topk_known_features": 32,
39
+ "unknown_topk": 128,
40
+ "use_unknown": true,
41
+ "apply_topk_to_unknown": true,
42
+ "topk_on_logits": false,
43
+ "factorize_unknown": true,
44
+ "factorize_rank": 256,
45
+ "use_epsilon_correction": true,
46
+ "block_size": 4096,
47
+ "pad_multiple": 16,
48
+ "store_unknown_weights": false,
49
+ "inject_layer": 16,
50
+ "inject_alpha": 1.0
51
+ }
52
+ }
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c273e89844fece6f6b8aa8fb00a1a6d2f180bf6e207c823e96d9531e405fbd0
3
+ size 5200533176
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:767d142d8ef12520891a8cc229e4e9fd845984e7daf5208f5dcb0276e3a6d2b1
3
+ size 5259853080
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:141fd4be20aba9caf76534d16b1cc81037d348e6f3a29a4358a93414c2545035
3
+ size 5259853096
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29136b4ed062ab4d9c24b081f27000a5eff35806aa5de47d73208629ba9498e9
3
+ size 1063347184
model.safetensors.index.json ADDED
@@ -0,0 +1,271 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 16783556608
4
+ },
5
+ "weight_map": {
6
+ "transformer.tok_emb.weight": "model-00001-of-00004.safetensors",
7
+ "transformer.blocks.0.ln_1.weight": "model-00001-of-00004.safetensors",
8
+ "transformer.blocks.0.ln_2.weight": "model-00001-of-00004.safetensors",
9
+ "transformer.blocks.0.attn.c_attn.weight": "model-00001-of-00004.safetensors",
10
+ "transformer.blocks.0.attn.c_proj.weight": "model-00001-of-00004.safetensors",
11
+ "transformer.blocks.0.attn.q_norm.weight": "model-00001-of-00004.safetensors",
12
+ "transformer.blocks.0.attn.k_norm.weight": "model-00001-of-00004.safetensors",
13
+ "transformer.blocks.0.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
14
+ "transformer.blocks.0.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
15
+ "transformer.blocks.1.ln_1.weight": "model-00001-of-00004.safetensors",
16
+ "transformer.blocks.1.ln_2.weight": "model-00001-of-00004.safetensors",
17
+ "transformer.blocks.1.attn.c_attn.weight": "model-00001-of-00004.safetensors",
18
+ "transformer.blocks.1.attn.c_proj.weight": "model-00001-of-00004.safetensors",
19
+ "transformer.blocks.1.attn.q_norm.weight": "model-00001-of-00004.safetensors",
20
+ "transformer.blocks.1.attn.k_norm.weight": "model-00001-of-00004.safetensors",
21
+ "transformer.blocks.1.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
22
+ "transformer.blocks.1.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
23
+ "transformer.blocks.2.ln_1.weight": "model-00001-of-00004.safetensors",
24
+ "transformer.blocks.2.ln_2.weight": "model-00001-of-00004.safetensors",
25
+ "transformer.blocks.2.attn.c_attn.weight": "model-00001-of-00004.safetensors",
26
+ "transformer.blocks.2.attn.c_proj.weight": "model-00001-of-00004.safetensors",
27
+ "transformer.blocks.2.attn.q_norm.weight": "model-00001-of-00004.safetensors",
28
+ "transformer.blocks.2.attn.k_norm.weight": "model-00001-of-00004.safetensors",
29
+ "transformer.blocks.2.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
30
+ "transformer.blocks.2.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
31
+ "transformer.blocks.3.ln_1.weight": "model-00001-of-00004.safetensors",
32
+ "transformer.blocks.3.ln_2.weight": "model-00001-of-00004.safetensors",
33
+ "transformer.blocks.3.attn.c_attn.weight": "model-00001-of-00004.safetensors",
34
+ "transformer.blocks.3.attn.c_proj.weight": "model-00001-of-00004.safetensors",
35
+ "transformer.blocks.3.attn.q_norm.weight": "model-00001-of-00004.safetensors",
36
+ "transformer.blocks.3.attn.k_norm.weight": "model-00001-of-00004.safetensors",
37
+ "transformer.blocks.3.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
38
+ "transformer.blocks.3.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
39
+ "transformer.blocks.4.ln_1.weight": "model-00001-of-00004.safetensors",
40
+ "transformer.blocks.4.ln_2.weight": "model-00001-of-00004.safetensors",
41
+ "transformer.blocks.4.attn.c_attn.weight": "model-00001-of-00004.safetensors",
42
+ "transformer.blocks.4.attn.c_proj.weight": "model-00001-of-00004.safetensors",
43
+ "transformer.blocks.4.attn.q_norm.weight": "model-00001-of-00004.safetensors",
44
+ "transformer.blocks.4.attn.k_norm.weight": "model-00001-of-00004.safetensors",
45
+ "transformer.blocks.4.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
46
+ "transformer.blocks.4.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
47
+ "transformer.blocks.5.ln_1.weight": "model-00001-of-00004.safetensors",
48
+ "transformer.blocks.5.ln_2.weight": "model-00001-of-00004.safetensors",
49
+ "transformer.blocks.5.attn.c_attn.weight": "model-00001-of-00004.safetensors",
50
+ "transformer.blocks.5.attn.c_proj.weight": "model-00001-of-00004.safetensors",
51
+ "transformer.blocks.5.attn.q_norm.weight": "model-00001-of-00004.safetensors",
52
+ "transformer.blocks.5.attn.k_norm.weight": "model-00001-of-00004.safetensors",
53
+ "transformer.blocks.5.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
54
+ "transformer.blocks.5.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
55
+ "transformer.blocks.6.ln_1.weight": "model-00001-of-00004.safetensors",
56
+ "transformer.blocks.6.ln_2.weight": "model-00001-of-00004.safetensors",
57
+ "transformer.blocks.6.attn.c_attn.weight": "model-00001-of-00004.safetensors",
58
+ "transformer.blocks.6.attn.c_proj.weight": "model-00001-of-00004.safetensors",
59
+ "transformer.blocks.6.attn.q_norm.weight": "model-00001-of-00004.safetensors",
60
+ "transformer.blocks.6.attn.k_norm.weight": "model-00001-of-00004.safetensors",
61
+ "transformer.blocks.6.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
62
+ "transformer.blocks.6.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
63
+ "transformer.blocks.7.ln_1.weight": "model-00001-of-00004.safetensors",
64
+ "transformer.blocks.7.ln_2.weight": "model-00001-of-00004.safetensors",
65
+ "transformer.blocks.7.attn.c_attn.weight": "model-00001-of-00004.safetensors",
66
+ "transformer.blocks.7.attn.c_proj.weight": "model-00001-of-00004.safetensors",
67
+ "transformer.blocks.7.attn.q_norm.weight": "model-00001-of-00004.safetensors",
68
+ "transformer.blocks.7.attn.k_norm.weight": "model-00001-of-00004.safetensors",
69
+ "transformer.blocks.7.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
70
+ "transformer.blocks.7.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
71
+ "transformer.blocks.8.ln_1.weight": "model-00001-of-00004.safetensors",
72
+ "transformer.blocks.8.ln_2.weight": "model-00001-of-00004.safetensors",
73
+ "transformer.blocks.8.attn.c_attn.weight": "model-00001-of-00004.safetensors",
74
+ "transformer.blocks.8.attn.c_proj.weight": "model-00001-of-00004.safetensors",
75
+ "transformer.blocks.8.attn.q_norm.weight": "model-00001-of-00004.safetensors",
76
+ "transformer.blocks.8.attn.k_norm.weight": "model-00001-of-00004.safetensors",
77
+ "transformer.blocks.8.mlp.c_fc.weight": "model-00001-of-00004.safetensors",
78
+ "transformer.blocks.8.mlp.c_proj.weight": "model-00001-of-00004.safetensors",
79
+ "transformer.blocks.9.ln_1.weight": "model-00001-of-00004.safetensors",
80
+ "transformer.blocks.9.ln_2.weight": "model-00001-of-00004.safetensors",
81
+ "transformer.blocks.9.attn.c_attn.weight": "model-00001-of-00004.safetensors",
82
+ "transformer.blocks.9.attn.c_proj.weight": "model-00001-of-00004.safetensors",
83
+ "transformer.blocks.9.attn.q_norm.weight": "model-00001-of-00004.safetensors",
84
+ "transformer.blocks.9.attn.k_norm.weight": "model-00001-of-00004.safetensors",
85
+ "transformer.blocks.9.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
86
+ "transformer.blocks.9.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
87
+ "transformer.blocks.10.ln_1.weight": "model-00002-of-00004.safetensors",
88
+ "transformer.blocks.10.ln_2.weight": "model-00002-of-00004.safetensors",
89
+ "transformer.blocks.10.attn.c_attn.weight": "model-00002-of-00004.safetensors",
90
+ "transformer.blocks.10.attn.c_proj.weight": "model-00002-of-00004.safetensors",
91
+ "transformer.blocks.10.attn.q_norm.weight": "model-00002-of-00004.safetensors",
92
+ "transformer.blocks.10.attn.k_norm.weight": "model-00002-of-00004.safetensors",
93
+ "transformer.blocks.10.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
94
+ "transformer.blocks.10.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
95
+ "transformer.blocks.11.ln_1.weight": "model-00002-of-00004.safetensors",
96
+ "transformer.blocks.11.ln_2.weight": "model-00002-of-00004.safetensors",
97
+ "transformer.blocks.11.attn.c_attn.weight": "model-00002-of-00004.safetensors",
98
+ "transformer.blocks.11.attn.c_proj.weight": "model-00002-of-00004.safetensors",
99
+ "transformer.blocks.11.attn.q_norm.weight": "model-00002-of-00004.safetensors",
100
+ "transformer.blocks.11.attn.k_norm.weight": "model-00002-of-00004.safetensors",
101
+ "transformer.blocks.11.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
102
+ "transformer.blocks.11.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
103
+ "transformer.blocks.12.ln_1.weight": "model-00002-of-00004.safetensors",
104
+ "transformer.blocks.12.ln_2.weight": "model-00002-of-00004.safetensors",
105
+ "transformer.blocks.12.attn.c_attn.weight": "model-00002-of-00004.safetensors",
106
+ "transformer.blocks.12.attn.c_proj.weight": "model-00002-of-00004.safetensors",
107
+ "transformer.blocks.12.attn.q_norm.weight": "model-00002-of-00004.safetensors",
108
+ "transformer.blocks.12.attn.k_norm.weight": "model-00002-of-00004.safetensors",
109
+ "transformer.blocks.12.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
110
+ "transformer.blocks.12.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
111
+ "transformer.blocks.13.ln_1.weight": "model-00002-of-00004.safetensors",
112
+ "transformer.blocks.13.ln_2.weight": "model-00002-of-00004.safetensors",
113
+ "transformer.blocks.13.attn.c_attn.weight": "model-00002-of-00004.safetensors",
114
+ "transformer.blocks.13.attn.c_proj.weight": "model-00002-of-00004.safetensors",
115
+ "transformer.blocks.13.attn.q_norm.weight": "model-00002-of-00004.safetensors",
116
+ "transformer.blocks.13.attn.k_norm.weight": "model-00002-of-00004.safetensors",
117
+ "transformer.blocks.13.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
118
+ "transformer.blocks.13.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
119
+ "transformer.blocks.14.ln_1.weight": "model-00002-of-00004.safetensors",
120
+ "transformer.blocks.14.ln_2.weight": "model-00002-of-00004.safetensors",
121
+ "transformer.blocks.14.attn.c_attn.weight": "model-00002-of-00004.safetensors",
122
+ "transformer.blocks.14.attn.c_proj.weight": "model-00002-of-00004.safetensors",
123
+ "transformer.blocks.14.attn.q_norm.weight": "model-00002-of-00004.safetensors",
124
+ "transformer.blocks.14.attn.k_norm.weight": "model-00002-of-00004.safetensors",
125
+ "transformer.blocks.14.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
126
+ "transformer.blocks.14.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
127
+ "transformer.blocks.15.ln_1.weight": "model-00002-of-00004.safetensors",
128
+ "transformer.blocks.15.ln_2.weight": "model-00002-of-00004.safetensors",
129
+ "transformer.blocks.15.attn.c_attn.weight": "model-00002-of-00004.safetensors",
130
+ "transformer.blocks.15.attn.c_proj.weight": "model-00002-of-00004.safetensors",
131
+ "transformer.blocks.15.attn.q_norm.weight": "model-00002-of-00004.safetensors",
132
+ "transformer.blocks.15.attn.k_norm.weight": "model-00002-of-00004.safetensors",
133
+ "transformer.blocks.15.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
134
+ "transformer.blocks.15.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
135
+ "transformer.blocks.16.ln_1.weight": "model-00002-of-00004.safetensors",
136
+ "transformer.blocks.16.ln_2.weight": "model-00002-of-00004.safetensors",
137
+ "transformer.blocks.16.attn.c_attn.weight": "model-00002-of-00004.safetensors",
138
+ "transformer.blocks.16.attn.c_proj.weight": "model-00002-of-00004.safetensors",
139
+ "transformer.blocks.16.attn.q_norm.weight": "model-00002-of-00004.safetensors",
140
+ "transformer.blocks.16.attn.k_norm.weight": "model-00002-of-00004.safetensors",
141
+ "transformer.blocks.16.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
142
+ "transformer.blocks.16.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
143
+ "transformer.blocks.17.ln_1.weight": "model-00002-of-00004.safetensors",
144
+ "transformer.blocks.17.ln_2.weight": "model-00002-of-00004.safetensors",
145
+ "transformer.blocks.17.attn.c_attn.weight": "model-00002-of-00004.safetensors",
146
+ "transformer.blocks.17.attn.c_proj.weight": "model-00002-of-00004.safetensors",
147
+ "transformer.blocks.17.attn.q_norm.weight": "model-00002-of-00004.safetensors",
148
+ "transformer.blocks.17.attn.k_norm.weight": "model-00002-of-00004.safetensors",
149
+ "transformer.blocks.17.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
150
+ "transformer.blocks.17.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
151
+ "transformer.blocks.18.ln_1.weight": "model-00002-of-00004.safetensors",
152
+ "transformer.blocks.18.ln_2.weight": "model-00002-of-00004.safetensors",
153
+ "transformer.blocks.18.attn.c_attn.weight": "model-00002-of-00004.safetensors",
154
+ "transformer.blocks.18.attn.c_proj.weight": "model-00002-of-00004.safetensors",
155
+ "transformer.blocks.18.attn.q_norm.weight": "model-00002-of-00004.safetensors",
156
+ "transformer.blocks.18.attn.k_norm.weight": "model-00002-of-00004.safetensors",
157
+ "transformer.blocks.18.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
158
+ "transformer.blocks.18.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
159
+ "transformer.blocks.19.ln_1.weight": "model-00002-of-00004.safetensors",
160
+ "transformer.blocks.19.ln_2.weight": "model-00002-of-00004.safetensors",
161
+ "transformer.blocks.19.attn.c_attn.weight": "model-00002-of-00004.safetensors",
162
+ "transformer.blocks.19.attn.c_proj.weight": "model-00002-of-00004.safetensors",
163
+ "transformer.blocks.19.attn.q_norm.weight": "model-00002-of-00004.safetensors",
164
+ "transformer.blocks.19.attn.k_norm.weight": "model-00002-of-00004.safetensors",
165
+ "transformer.blocks.19.mlp.c_fc.weight": "model-00002-of-00004.safetensors",
166
+ "transformer.blocks.19.mlp.c_proj.weight": "model-00002-of-00004.safetensors",
167
+ "transformer.blocks.20.ln_1.weight": "model-00002-of-00004.safetensors",
168
+ "transformer.blocks.20.ln_2.weight": "model-00002-of-00004.safetensors",
169
+ "transformer.blocks.20.attn.c_attn.weight": "model-00002-of-00004.safetensors",
170
+ "transformer.blocks.20.attn.c_proj.weight": "model-00002-of-00004.safetensors",
171
+ "transformer.blocks.20.attn.q_norm.weight": "model-00002-of-00004.safetensors",
172
+ "transformer.blocks.20.attn.k_norm.weight": "model-00002-of-00004.safetensors",
173
+ "transformer.blocks.20.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
174
+ "transformer.blocks.20.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
175
+ "transformer.blocks.21.ln_1.weight": "model-00003-of-00004.safetensors",
176
+ "transformer.blocks.21.ln_2.weight": "model-00003-of-00004.safetensors",
177
+ "transformer.blocks.21.attn.c_attn.weight": "model-00003-of-00004.safetensors",
178
+ "transformer.blocks.21.attn.c_proj.weight": "model-00003-of-00004.safetensors",
179
+ "transformer.blocks.21.attn.q_norm.weight": "model-00003-of-00004.safetensors",
180
+ "transformer.blocks.21.attn.k_norm.weight": "model-00003-of-00004.safetensors",
181
+ "transformer.blocks.21.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
182
+ "transformer.blocks.21.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
183
+ "transformer.blocks.22.ln_1.weight": "model-00003-of-00004.safetensors",
184
+ "transformer.blocks.22.ln_2.weight": "model-00003-of-00004.safetensors",
185
+ "transformer.blocks.22.attn.c_attn.weight": "model-00003-of-00004.safetensors",
186
+ "transformer.blocks.22.attn.c_proj.weight": "model-00003-of-00004.safetensors",
187
+ "transformer.blocks.22.attn.q_norm.weight": "model-00003-of-00004.safetensors",
188
+ "transformer.blocks.22.attn.k_norm.weight": "model-00003-of-00004.safetensors",
189
+ "transformer.blocks.22.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
190
+ "transformer.blocks.22.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
191
+ "transformer.blocks.23.ln_1.weight": "model-00003-of-00004.safetensors",
192
+ "transformer.blocks.23.ln_2.weight": "model-00003-of-00004.safetensors",
193
+ "transformer.blocks.23.attn.c_attn.weight": "model-00003-of-00004.safetensors",
194
+ "transformer.blocks.23.attn.c_proj.weight": "model-00003-of-00004.safetensors",
195
+ "transformer.blocks.23.attn.q_norm.weight": "model-00003-of-00004.safetensors",
196
+ "transformer.blocks.23.attn.k_norm.weight": "model-00003-of-00004.safetensors",
197
+ "transformer.blocks.23.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
198
+ "transformer.blocks.23.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
199
+ "transformer.blocks.24.ln_1.weight": "model-00003-of-00004.safetensors",
200
+ "transformer.blocks.24.ln_2.weight": "model-00003-of-00004.safetensors",
201
+ "transformer.blocks.24.attn.c_attn.weight": "model-00003-of-00004.safetensors",
202
+ "transformer.blocks.24.attn.c_proj.weight": "model-00003-of-00004.safetensors",
203
+ "transformer.blocks.24.attn.q_norm.weight": "model-00003-of-00004.safetensors",
204
+ "transformer.blocks.24.attn.k_norm.weight": "model-00003-of-00004.safetensors",
205
+ "transformer.blocks.24.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
206
+ "transformer.blocks.24.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
207
+ "transformer.blocks.25.ln_1.weight": "model-00003-of-00004.safetensors",
208
+ "transformer.blocks.25.ln_2.weight": "model-00003-of-00004.safetensors",
209
+ "transformer.blocks.25.attn.c_attn.weight": "model-00003-of-00004.safetensors",
210
+ "transformer.blocks.25.attn.c_proj.weight": "model-00003-of-00004.safetensors",
211
+ "transformer.blocks.25.attn.q_norm.weight": "model-00003-of-00004.safetensors",
212
+ "transformer.blocks.25.attn.k_norm.weight": "model-00003-of-00004.safetensors",
213
+ "transformer.blocks.25.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
214
+ "transformer.blocks.25.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
215
+ "transformer.blocks.26.ln_1.weight": "model-00003-of-00004.safetensors",
216
+ "transformer.blocks.26.ln_2.weight": "model-00003-of-00004.safetensors",
217
+ "transformer.blocks.26.attn.c_attn.weight": "model-00003-of-00004.safetensors",
218
+ "transformer.blocks.26.attn.c_proj.weight": "model-00003-of-00004.safetensors",
219
+ "transformer.blocks.26.attn.q_norm.weight": "model-00003-of-00004.safetensors",
220
+ "transformer.blocks.26.attn.k_norm.weight": "model-00003-of-00004.safetensors",
221
+ "transformer.blocks.26.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
222
+ "transformer.blocks.26.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
223
+ "transformer.blocks.27.ln_1.weight": "model-00003-of-00004.safetensors",
224
+ "transformer.blocks.27.ln_2.weight": "model-00003-of-00004.safetensors",
225
+ "transformer.blocks.27.attn.c_attn.weight": "model-00003-of-00004.safetensors",
226
+ "transformer.blocks.27.attn.c_proj.weight": "model-00003-of-00004.safetensors",
227
+ "transformer.blocks.27.attn.q_norm.weight": "model-00003-of-00004.safetensors",
228
+ "transformer.blocks.27.attn.k_norm.weight": "model-00003-of-00004.safetensors",
229
+ "transformer.blocks.27.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
230
+ "transformer.blocks.27.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
231
+ "transformer.blocks.28.ln_1.weight": "model-00003-of-00004.safetensors",
232
+ "transformer.blocks.28.ln_2.weight": "model-00003-of-00004.safetensors",
233
+ "transformer.blocks.28.attn.c_attn.weight": "model-00003-of-00004.safetensors",
234
+ "transformer.blocks.28.attn.c_proj.weight": "model-00003-of-00004.safetensors",
235
+ "transformer.blocks.28.attn.q_norm.weight": "model-00003-of-00004.safetensors",
236
+ "transformer.blocks.28.attn.k_norm.weight": "model-00003-of-00004.safetensors",
237
+ "transformer.blocks.28.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
238
+ "transformer.blocks.28.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
239
+ "transformer.blocks.29.ln_1.weight": "model-00003-of-00004.safetensors",
240
+ "transformer.blocks.29.ln_2.weight": "model-00003-of-00004.safetensors",
241
+ "transformer.blocks.29.attn.c_attn.weight": "model-00003-of-00004.safetensors",
242
+ "transformer.blocks.29.attn.c_proj.weight": "model-00003-of-00004.safetensors",
243
+ "transformer.blocks.29.attn.q_norm.weight": "model-00003-of-00004.safetensors",
244
+ "transformer.blocks.29.attn.k_norm.weight": "model-00003-of-00004.safetensors",
245
+ "transformer.blocks.29.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
246
+ "transformer.blocks.29.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
247
+ "transformer.blocks.30.ln_1.weight": "model-00003-of-00004.safetensors",
248
+ "transformer.blocks.30.ln_2.weight": "model-00003-of-00004.safetensors",
249
+ "transformer.blocks.30.attn.c_attn.weight": "model-00003-of-00004.safetensors",
250
+ "transformer.blocks.30.attn.c_proj.weight": "model-00003-of-00004.safetensors",
251
+ "transformer.blocks.30.attn.q_norm.weight": "model-00003-of-00004.safetensors",
252
+ "transformer.blocks.30.attn.k_norm.weight": "model-00003-of-00004.safetensors",
253
+ "transformer.blocks.30.mlp.c_fc.weight": "model-00003-of-00004.safetensors",
254
+ "transformer.blocks.30.mlp.c_proj.weight": "model-00003-of-00004.safetensors",
255
+ "transformer.blocks.31.ln_1.weight": "model-00003-of-00004.safetensors",
256
+ "transformer.blocks.31.ln_2.weight": "model-00003-of-00004.safetensors",
257
+ "transformer.blocks.31.attn.c_attn.weight": "model-00003-of-00004.safetensors",
258
+ "transformer.blocks.31.attn.c_proj.weight": "model-00003-of-00004.safetensors",
259
+ "transformer.blocks.31.attn.q_norm.weight": "model-00003-of-00004.safetensors",
260
+ "transformer.blocks.31.attn.k_norm.weight": "model-00003-of-00004.safetensors",
261
+ "transformer.blocks.31.mlp.c_fc.weight": "model-00004-of-00004.safetensors",
262
+ "transformer.blocks.31.mlp.c_proj.weight": "model-00004-of-00004.safetensors",
263
+ "transformer.ln_f.weight": "model-00004-of-00004.safetensors",
264
+ "known_head.concept_embedding.weight": "model-00004-of-00004.safetensors",
265
+ "known_head.concept_predictor.weight": "model-00004-of-00004.safetensors",
266
+ "unknown_head.embedding_coef.weight": "model-00004-of-00004.safetensors",
267
+ "unknown_head.embedding_basis.weight": "model-00004-of-00004.safetensors",
268
+ "unknown_head.predictor_down.weight": "model-00004-of-00004.safetensors",
269
+ "unknown_head.predictor_up.weight": "model-00004-of-00004.safetensors"
270
+ }
271
+ }