AndrewYan commited on
Commit
0de9815
·
verified ·
1 Parent(s): 086e69a

Upload model

Browse files
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/mnt/workspace/32b",
3
+ "architectures": [
4
+ "Qwen2Model"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 5120,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 27648,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 70,
15
+ "model_type": "qwen2",
16
+ "num_attention_heads": 40,
17
+ "num_hidden_layers": 64,
18
+ "num_key_value_heads": 8,
19
+ "rms_norm_eps": 1e-06,
20
+ "rope_scaling": null,
21
+ "rope_theta": 1000000.0,
22
+ "sliding_window": null,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.48.3",
26
+ "use_cache": false,
27
+ "use_sliding_window": false,
28
+ "vocab_size": 152064
29
+ }
model-00001-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3757841452a1ff455193537c8ab25508560590fae0110eb1b1e7a300a1ce98f
3
+ size 4498420808
model-00002-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbb717f05e9f50ce029fe60ecce20f66d18ce2ce996fa97dae72cf19f60b3afd
3
+ size 4718804568
model-00003-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6576199ae6bb08bc648e495e713b2dd21111e1fb29a3aa879e0407a55e2f677
3
+ size 4467075728
model-00004-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9af77d0f9bc07e9d5b7d1d3b1803a907b5dbcfcb331d259bf568a9509ae93d0c
3
+ size 4467075728
model-00005-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62ada162cf47edf125e41a11417ff7ecc2455c2549a652395e98c31f32488049
3
+ size 4718804552
model-00006-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60220589b80f83eeb500f3cd2bfe32b4192f786ec70ade61661e74bdc4883af7
3
+ size 4467075752
model-00007-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64dfd138293999f5292d1ee307e178b16f6c77cdbb8fc98508c4ae7278e4ba1f
3
+ size 4467075752
model-00008-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0819d756dc60d18c46020759381266ac31edf0c8e7a9e9e8d931afd719094d1
3
+ size 4718804600
model-00009-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a496d37f56af9949adc70c7e5b8273fced013cb1e0c75d168170c03db62036d
3
+ size 4467075752
model-00010-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c63ce7531264d8fbc26e9087541dbe4bbe102dabcf3a75aa4824cadc09968321
3
+ size 4467075752
model-00011-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ae094bc3e285f6a235a8ac22e8755f62d627f2c3f1255c2889238d692105e2d
3
+ size 4718804600
model-00012-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:290b3448cb206ce54356b7756f5c82dc83c0290cdfee9dfe17c527117a4d1045
3
+ size 4467075752
model-00013-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:150d59413ed6c2e9f0a5efe74d22015616e0e69d2efd75c13dfe0be1d200eda8
3
+ size 4467075752
model-00014-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57fd7902b336b21ac822d9ddf2c9bceef42d86bde474314bc5bb4a89e086cac9
3
+ size 4718804600
model-00015-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f32d932b200f4743a6dd2805584355439d4cd10a20892ced5777a6fd2ba57a1f
3
+ size 4467075752
model-00016-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d4575dc45209afdac7d1719a9098ac84c754434becc18d17b2c6cb08d418340
3
+ size 4467075752
model-00017-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5057c6907e44839cbfa9b547aeae526e2a87cb002838dabe8ac9dfc6bd61a668
3
+ size 4718804600
model-00018-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c105aa72ec2081192c38ddb17b3523561d33731435da6c2124ca05b2acd05c3
3
+ size 4467075752
model-00019-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:673d68c4eaf9fe055a0fda1e1e91a27afa79f367e7a7fca0fdd0b28eb7098e14
3
+ size 4467075752
model-00020-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49b9df2c4f2b126b9416e8747624740619fb6e2974979f984950e818883113e8
3
+ size 4718804600
model-00021-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbb18593421db5857a3b43e898ad0109f226446e1ef2777a03db4d9c66ecf361
3
+ size 4467075752
model-00022-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a1339d92a51d023317639c81fde584b7a14a1ba933ab0ff01b7fb293b8991d7
3
+ size 4467075752
model-00023-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23e13406acde194acfa96f32c6d62acc0bfbe98003cbcf27ca7deaea7b7c2d70
3
+ size 4718804600
model-00024-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89a71bb01d6b8b1dc294c3a8f015eae796febad62841eb2149508d684b8f2062
3
+ size 4467075752
model-00025-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bd91d534f4ca4cfb46cd3b9839be3a2a096a8ed09a8c1f376019693b25810b9
3
+ size 4467075752
model-00026-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bf0cac9d2f286228673941f5f3c7fc2a5320843fa8e9fa95d0335ba4ab744d2
3
+ size 4718804600
model-00027-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88db274b376f9a9152beffe11ed3adbfee2a83ccf8d4c9a12ca2a21944219015
3
+ size 4467075752
model-00028-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:254871a1c2bb9e18697b001441ac780dfb63f99e6e127fe166d28bbc3db70602
3
+ size 4467075752
model-00029-of-00029.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18b15a6c5cc7fe9bd4266162c99723a218e7bf914217ead194b1b970475e0b1d
3
+ size 566292912
model.safetensors.index.json ADDED
@@ -0,0 +1,777 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 127941234688
4
+ },
5
+ "weight_map": {
6
+ "embed_tokens.weight": "model-00001-of-00029.safetensors",
7
+ "layers.0.input_layernorm.weight": "model-00002-of-00029.safetensors",
8
+ "layers.0.mlp.down_proj.weight": "model-00002-of-00029.safetensors",
9
+ "layers.0.mlp.gate_proj.weight": "model-00001-of-00029.safetensors",
10
+ "layers.0.mlp.up_proj.weight": "model-00001-of-00029.safetensors",
11
+ "layers.0.post_attention_layernorm.weight": "model-00002-of-00029.safetensors",
12
+ "layers.0.self_attn.k_proj.bias": "model-00001-of-00029.safetensors",
13
+ "layers.0.self_attn.k_proj.weight": "model-00001-of-00029.safetensors",
14
+ "layers.0.self_attn.o_proj.weight": "model-00001-of-00029.safetensors",
15
+ "layers.0.self_attn.q_proj.bias": "model-00001-of-00029.safetensors",
16
+ "layers.0.self_attn.q_proj.weight": "model-00001-of-00029.safetensors",
17
+ "layers.0.self_attn.v_proj.bias": "model-00001-of-00029.safetensors",
18
+ "layers.0.self_attn.v_proj.weight": "model-00001-of-00029.safetensors",
19
+ "layers.1.input_layernorm.weight": "model-00002-of-00029.safetensors",
20
+ "layers.1.mlp.down_proj.weight": "model-00002-of-00029.safetensors",
21
+ "layers.1.mlp.gate_proj.weight": "model-00002-of-00029.safetensors",
22
+ "layers.1.mlp.up_proj.weight": "model-00002-of-00029.safetensors",
23
+ "layers.1.post_attention_layernorm.weight": "model-00002-of-00029.safetensors",
24
+ "layers.1.self_attn.k_proj.bias": "model-00002-of-00029.safetensors",
25
+ "layers.1.self_attn.k_proj.weight": "model-00002-of-00029.safetensors",
26
+ "layers.1.self_attn.o_proj.weight": "model-00002-of-00029.safetensors",
27
+ "layers.1.self_attn.q_proj.bias": "model-00002-of-00029.safetensors",
28
+ "layers.1.self_attn.q_proj.weight": "model-00002-of-00029.safetensors",
29
+ "layers.1.self_attn.v_proj.bias": "model-00002-of-00029.safetensors",
30
+ "layers.1.self_attn.v_proj.weight": "model-00002-of-00029.safetensors",
31
+ "layers.10.input_layernorm.weight": "model-00006-of-00029.safetensors",
32
+ "layers.10.mlp.down_proj.weight": "model-00006-of-00029.safetensors",
33
+ "layers.10.mlp.gate_proj.weight": "model-00006-of-00029.safetensors",
34
+ "layers.10.mlp.up_proj.weight": "model-00006-of-00029.safetensors",
35
+ "layers.10.post_attention_layernorm.weight": "model-00006-of-00029.safetensors",
36
+ "layers.10.self_attn.k_proj.bias": "model-00005-of-00029.safetensors",
37
+ "layers.10.self_attn.k_proj.weight": "model-00005-of-00029.safetensors",
38
+ "layers.10.self_attn.o_proj.weight": "model-00005-of-00029.safetensors",
39
+ "layers.10.self_attn.q_proj.bias": "model-00005-of-00029.safetensors",
40
+ "layers.10.self_attn.q_proj.weight": "model-00005-of-00029.safetensors",
41
+ "layers.10.self_attn.v_proj.bias": "model-00005-of-00029.safetensors",
42
+ "layers.10.self_attn.v_proj.weight": "model-00005-of-00029.safetensors",
43
+ "layers.11.input_layernorm.weight": "model-00006-of-00029.safetensors",
44
+ "layers.11.mlp.down_proj.weight": "model-00006-of-00029.safetensors",
45
+ "layers.11.mlp.gate_proj.weight": "model-00006-of-00029.safetensors",
46
+ "layers.11.mlp.up_proj.weight": "model-00006-of-00029.safetensors",
47
+ "layers.11.post_attention_layernorm.weight": "model-00006-of-00029.safetensors",
48
+ "layers.11.self_attn.k_proj.bias": "model-00006-of-00029.safetensors",
49
+ "layers.11.self_attn.k_proj.weight": "model-00006-of-00029.safetensors",
50
+ "layers.11.self_attn.o_proj.weight": "model-00006-of-00029.safetensors",
51
+ "layers.11.self_attn.q_proj.bias": "model-00006-of-00029.safetensors",
52
+ "layers.11.self_attn.q_proj.weight": "model-00006-of-00029.safetensors",
53
+ "layers.11.self_attn.v_proj.bias": "model-00006-of-00029.safetensors",
54
+ "layers.11.self_attn.v_proj.weight": "model-00006-of-00029.safetensors",
55
+ "layers.12.input_layernorm.weight": "model-00007-of-00029.safetensors",
56
+ "layers.12.mlp.down_proj.weight": "model-00007-of-00029.safetensors",
57
+ "layers.12.mlp.gate_proj.weight": "model-00006-of-00029.safetensors",
58
+ "layers.12.mlp.up_proj.weight": "model-00007-of-00029.safetensors",
59
+ "layers.12.post_attention_layernorm.weight": "model-00007-of-00029.safetensors",
60
+ "layers.12.self_attn.k_proj.bias": "model-00006-of-00029.safetensors",
61
+ "layers.12.self_attn.k_proj.weight": "model-00006-of-00029.safetensors",
62
+ "layers.12.self_attn.o_proj.weight": "model-00006-of-00029.safetensors",
63
+ "layers.12.self_attn.q_proj.bias": "model-00006-of-00029.safetensors",
64
+ "layers.12.self_attn.q_proj.weight": "model-00006-of-00029.safetensors",
65
+ "layers.12.self_attn.v_proj.bias": "model-00006-of-00029.safetensors",
66
+ "layers.12.self_attn.v_proj.weight": "model-00006-of-00029.safetensors",
67
+ "layers.13.input_layernorm.weight": "model-00007-of-00029.safetensors",
68
+ "layers.13.mlp.down_proj.weight": "model-00007-of-00029.safetensors",
69
+ "layers.13.mlp.gate_proj.weight": "model-00007-of-00029.safetensors",
70
+ "layers.13.mlp.up_proj.weight": "model-00007-of-00029.safetensors",
71
+ "layers.13.post_attention_layernorm.weight": "model-00007-of-00029.safetensors",
72
+ "layers.13.self_attn.k_proj.bias": "model-00007-of-00029.safetensors",
73
+ "layers.13.self_attn.k_proj.weight": "model-00007-of-00029.safetensors",
74
+ "layers.13.self_attn.o_proj.weight": "model-00007-of-00029.safetensors",
75
+ "layers.13.self_attn.q_proj.bias": "model-00007-of-00029.safetensors",
76
+ "layers.13.self_attn.q_proj.weight": "model-00007-of-00029.safetensors",
77
+ "layers.13.self_attn.v_proj.bias": "model-00007-of-00029.safetensors",
78
+ "layers.13.self_attn.v_proj.weight": "model-00007-of-00029.safetensors",
79
+ "layers.14.input_layernorm.weight": "model-00008-of-00029.safetensors",
80
+ "layers.14.mlp.down_proj.weight": "model-00008-of-00029.safetensors",
81
+ "layers.14.mlp.gate_proj.weight": "model-00007-of-00029.safetensors",
82
+ "layers.14.mlp.up_proj.weight": "model-00007-of-00029.safetensors",
83
+ "layers.14.post_attention_layernorm.weight": "model-00008-of-00029.safetensors",
84
+ "layers.14.self_attn.k_proj.bias": "model-00007-of-00029.safetensors",
85
+ "layers.14.self_attn.k_proj.weight": "model-00007-of-00029.safetensors",
86
+ "layers.14.self_attn.o_proj.weight": "model-00007-of-00029.safetensors",
87
+ "layers.14.self_attn.q_proj.bias": "model-00007-of-00029.safetensors",
88
+ "layers.14.self_attn.q_proj.weight": "model-00007-of-00029.safetensors",
89
+ "layers.14.self_attn.v_proj.bias": "model-00007-of-00029.safetensors",
90
+ "layers.14.self_attn.v_proj.weight": "model-00007-of-00029.safetensors",
91
+ "layers.15.input_layernorm.weight": "model-00008-of-00029.safetensors",
92
+ "layers.15.mlp.down_proj.weight": "model-00008-of-00029.safetensors",
93
+ "layers.15.mlp.gate_proj.weight": "model-00008-of-00029.safetensors",
94
+ "layers.15.mlp.up_proj.weight": "model-00008-of-00029.safetensors",
95
+ "layers.15.post_attention_layernorm.weight": "model-00008-of-00029.safetensors",
96
+ "layers.15.self_attn.k_proj.bias": "model-00008-of-00029.safetensors",
97
+ "layers.15.self_attn.k_proj.weight": "model-00008-of-00029.safetensors",
98
+ "layers.15.self_attn.o_proj.weight": "model-00008-of-00029.safetensors",
99
+ "layers.15.self_attn.q_proj.bias": "model-00008-of-00029.safetensors",
100
+ "layers.15.self_attn.q_proj.weight": "model-00008-of-00029.safetensors",
101
+ "layers.15.self_attn.v_proj.bias": "model-00008-of-00029.safetensors",
102
+ "layers.15.self_attn.v_proj.weight": "model-00008-of-00029.safetensors",
103
+ "layers.16.input_layernorm.weight": "model-00008-of-00029.safetensors",
104
+ "layers.16.mlp.down_proj.weight": "model-00008-of-00029.safetensors",
105
+ "layers.16.mlp.gate_proj.weight": "model-00008-of-00029.safetensors",
106
+ "layers.16.mlp.up_proj.weight": "model-00008-of-00029.safetensors",
107
+ "layers.16.post_attention_layernorm.weight": "model-00008-of-00029.safetensors",
108
+ "layers.16.self_attn.k_proj.bias": "model-00008-of-00029.safetensors",
109
+ "layers.16.self_attn.k_proj.weight": "model-00008-of-00029.safetensors",
110
+ "layers.16.self_attn.o_proj.weight": "model-00008-of-00029.safetensors",
111
+ "layers.16.self_attn.q_proj.bias": "model-00008-of-00029.safetensors",
112
+ "layers.16.self_attn.q_proj.weight": "model-00008-of-00029.safetensors",
113
+ "layers.16.self_attn.v_proj.bias": "model-00008-of-00029.safetensors",
114
+ "layers.16.self_attn.v_proj.weight": "model-00008-of-00029.safetensors",
115
+ "layers.17.input_layernorm.weight": "model-00009-of-00029.safetensors",
116
+ "layers.17.mlp.down_proj.weight": "model-00009-of-00029.safetensors",
117
+ "layers.17.mlp.gate_proj.weight": "model-00009-of-00029.safetensors",
118
+ "layers.17.mlp.up_proj.weight": "model-00009-of-00029.safetensors",
119
+ "layers.17.post_attention_layernorm.weight": "model-00009-of-00029.safetensors",
120
+ "layers.17.self_attn.k_proj.bias": "model-00008-of-00029.safetensors",
121
+ "layers.17.self_attn.k_proj.weight": "model-00008-of-00029.safetensors",
122
+ "layers.17.self_attn.o_proj.weight": "model-00008-of-00029.safetensors",
123
+ "layers.17.self_attn.q_proj.bias": "model-00008-of-00029.safetensors",
124
+ "layers.17.self_attn.q_proj.weight": "model-00008-of-00029.safetensors",
125
+ "layers.17.self_attn.v_proj.bias": "model-00008-of-00029.safetensors",
126
+ "layers.17.self_attn.v_proj.weight": "model-00008-of-00029.safetensors",
127
+ "layers.18.input_layernorm.weight": "model-00009-of-00029.safetensors",
128
+ "layers.18.mlp.down_proj.weight": "model-00009-of-00029.safetensors",
129
+ "layers.18.mlp.gate_proj.weight": "model-00009-of-00029.safetensors",
130
+ "layers.18.mlp.up_proj.weight": "model-00009-of-00029.safetensors",
131
+ "layers.18.post_attention_layernorm.weight": "model-00009-of-00029.safetensors",
132
+ "layers.18.self_attn.k_proj.bias": "model-00009-of-00029.safetensors",
133
+ "layers.18.self_attn.k_proj.weight": "model-00009-of-00029.safetensors",
134
+ "layers.18.self_attn.o_proj.weight": "model-00009-of-00029.safetensors",
135
+ "layers.18.self_attn.q_proj.bias": "model-00009-of-00029.safetensors",
136
+ "layers.18.self_attn.q_proj.weight": "model-00009-of-00029.safetensors",
137
+ "layers.18.self_attn.v_proj.bias": "model-00009-of-00029.safetensors",
138
+ "layers.18.self_attn.v_proj.weight": "model-00009-of-00029.safetensors",
139
+ "layers.19.input_layernorm.weight": "model-00010-of-00029.safetensors",
140
+ "layers.19.mlp.down_proj.weight": "model-00010-of-00029.safetensors",
141
+ "layers.19.mlp.gate_proj.weight": "model-00009-of-00029.safetensors",
142
+ "layers.19.mlp.up_proj.weight": "model-00010-of-00029.safetensors",
143
+ "layers.19.post_attention_layernorm.weight": "model-00010-of-00029.safetensors",
144
+ "layers.19.self_attn.k_proj.bias": "model-00009-of-00029.safetensors",
145
+ "layers.19.self_attn.k_proj.weight": "model-00009-of-00029.safetensors",
146
+ "layers.19.self_attn.o_proj.weight": "model-00009-of-00029.safetensors",
147
+ "layers.19.self_attn.q_proj.bias": "model-00009-of-00029.safetensors",
148
+ "layers.19.self_attn.q_proj.weight": "model-00009-of-00029.safetensors",
149
+ "layers.19.self_attn.v_proj.bias": "model-00009-of-00029.safetensors",
150
+ "layers.19.self_attn.v_proj.weight": "model-00009-of-00029.safetensors",
151
+ "layers.2.input_layernorm.weight": "model-00002-of-00029.safetensors",
152
+ "layers.2.mlp.down_proj.weight": "model-00002-of-00029.safetensors",
153
+ "layers.2.mlp.gate_proj.weight": "model-00002-of-00029.safetensors",
154
+ "layers.2.mlp.up_proj.weight": "model-00002-of-00029.safetensors",
155
+ "layers.2.post_attention_layernorm.weight": "model-00002-of-00029.safetensors",
156
+ "layers.2.self_attn.k_proj.bias": "model-00002-of-00029.safetensors",
157
+ "layers.2.self_attn.k_proj.weight": "model-00002-of-00029.safetensors",
158
+ "layers.2.self_attn.o_proj.weight": "model-00002-of-00029.safetensors",
159
+ "layers.2.self_attn.q_proj.bias": "model-00002-of-00029.safetensors",
160
+ "layers.2.self_attn.q_proj.weight": "model-00002-of-00029.safetensors",
161
+ "layers.2.self_attn.v_proj.bias": "model-00002-of-00029.safetensors",
162
+ "layers.2.self_attn.v_proj.weight": "model-00002-of-00029.safetensors",
163
+ "layers.20.input_layernorm.weight": "model-00010-of-00029.safetensors",
164
+ "layers.20.mlp.down_proj.weight": "model-00010-of-00029.safetensors",
165
+ "layers.20.mlp.gate_proj.weight": "model-00010-of-00029.safetensors",
166
+ "layers.20.mlp.up_proj.weight": "model-00010-of-00029.safetensors",
167
+ "layers.20.post_attention_layernorm.weight": "model-00010-of-00029.safetensors",
168
+ "layers.20.self_attn.k_proj.bias": "model-00010-of-00029.safetensors",
169
+ "layers.20.self_attn.k_proj.weight": "model-00010-of-00029.safetensors",
170
+ "layers.20.self_attn.o_proj.weight": "model-00010-of-00029.safetensors",
171
+ "layers.20.self_attn.q_proj.bias": "model-00010-of-00029.safetensors",
172
+ "layers.20.self_attn.q_proj.weight": "model-00010-of-00029.safetensors",
173
+ "layers.20.self_attn.v_proj.bias": "model-00010-of-00029.safetensors",
174
+ "layers.20.self_attn.v_proj.weight": "model-00010-of-00029.safetensors",
175
+ "layers.21.input_layernorm.weight": "model-00011-of-00029.safetensors",
176
+ "layers.21.mlp.down_proj.weight": "model-00011-of-00029.safetensors",
177
+ "layers.21.mlp.gate_proj.weight": "model-00010-of-00029.safetensors",
178
+ "layers.21.mlp.up_proj.weight": "model-00010-of-00029.safetensors",
179
+ "layers.21.post_attention_layernorm.weight": "model-00011-of-00029.safetensors",
180
+ "layers.21.self_attn.k_proj.bias": "model-00010-of-00029.safetensors",
181
+ "layers.21.self_attn.k_proj.weight": "model-00010-of-00029.safetensors",
182
+ "layers.21.self_attn.o_proj.weight": "model-00010-of-00029.safetensors",
183
+ "layers.21.self_attn.q_proj.bias": "model-00010-of-00029.safetensors",
184
+ "layers.21.self_attn.q_proj.weight": "model-00010-of-00029.safetensors",
185
+ "layers.21.self_attn.v_proj.bias": "model-00010-of-00029.safetensors",
186
+ "layers.21.self_attn.v_proj.weight": "model-00010-of-00029.safetensors",
187
+ "layers.22.input_layernorm.weight": "model-00011-of-00029.safetensors",
188
+ "layers.22.mlp.down_proj.weight": "model-00011-of-00029.safetensors",
189
+ "layers.22.mlp.gate_proj.weight": "model-00011-of-00029.safetensors",
190
+ "layers.22.mlp.up_proj.weight": "model-00011-of-00029.safetensors",
191
+ "layers.22.post_attention_layernorm.weight": "model-00011-of-00029.safetensors",
192
+ "layers.22.self_attn.k_proj.bias": "model-00011-of-00029.safetensors",
193
+ "layers.22.self_attn.k_proj.weight": "model-00011-of-00029.safetensors",
194
+ "layers.22.self_attn.o_proj.weight": "model-00011-of-00029.safetensors",
195
+ "layers.22.self_attn.q_proj.bias": "model-00011-of-00029.safetensors",
196
+ "layers.22.self_attn.q_proj.weight": "model-00011-of-00029.safetensors",
197
+ "layers.22.self_attn.v_proj.bias": "model-00011-of-00029.safetensors",
198
+ "layers.22.self_attn.v_proj.weight": "model-00011-of-00029.safetensors",
199
+ "layers.23.input_layernorm.weight": "model-00011-of-00029.safetensors",
200
+ "layers.23.mlp.down_proj.weight": "model-00011-of-00029.safetensors",
201
+ "layers.23.mlp.gate_proj.weight": "model-00011-of-00029.safetensors",
202
+ "layers.23.mlp.up_proj.weight": "model-00011-of-00029.safetensors",
203
+ "layers.23.post_attention_layernorm.weight": "model-00011-of-00029.safetensors",
204
+ "layers.23.self_attn.k_proj.bias": "model-00011-of-00029.safetensors",
205
+ "layers.23.self_attn.k_proj.weight": "model-00011-of-00029.safetensors",
206
+ "layers.23.self_attn.o_proj.weight": "model-00011-of-00029.safetensors",
207
+ "layers.23.self_attn.q_proj.bias": "model-00011-of-00029.safetensors",
208
+ "layers.23.self_attn.q_proj.weight": "model-00011-of-00029.safetensors",
209
+ "layers.23.self_attn.v_proj.bias": "model-00011-of-00029.safetensors",
210
+ "layers.23.self_attn.v_proj.weight": "model-00011-of-00029.safetensors",
211
+ "layers.24.input_layernorm.weight": "model-00012-of-00029.safetensors",
212
+ "layers.24.mlp.down_proj.weight": "model-00012-of-00029.safetensors",
213
+ "layers.24.mlp.gate_proj.weight": "model-00012-of-00029.safetensors",
214
+ "layers.24.mlp.up_proj.weight": "model-00012-of-00029.safetensors",
215
+ "layers.24.post_attention_layernorm.weight": "model-00012-of-00029.safetensors",
216
+ "layers.24.self_attn.k_proj.bias": "model-00011-of-00029.safetensors",
217
+ "layers.24.self_attn.k_proj.weight": "model-00011-of-00029.safetensors",
218
+ "layers.24.self_attn.o_proj.weight": "model-00011-of-00029.safetensors",
219
+ "layers.24.self_attn.q_proj.bias": "model-00011-of-00029.safetensors",
220
+ "layers.24.self_attn.q_proj.weight": "model-00011-of-00029.safetensors",
221
+ "layers.24.self_attn.v_proj.bias": "model-00011-of-00029.safetensors",
222
+ "layers.24.self_attn.v_proj.weight": "model-00011-of-00029.safetensors",
223
+ "layers.25.input_layernorm.weight": "model-00012-of-00029.safetensors",
224
+ "layers.25.mlp.down_proj.weight": "model-00012-of-00029.safetensors",
225
+ "layers.25.mlp.gate_proj.weight": "model-00012-of-00029.safetensors",
226
+ "layers.25.mlp.up_proj.weight": "model-00012-of-00029.safetensors",
227
+ "layers.25.post_attention_layernorm.weight": "model-00012-of-00029.safetensors",
228
+ "layers.25.self_attn.k_proj.bias": "model-00012-of-00029.safetensors",
229
+ "layers.25.self_attn.k_proj.weight": "model-00012-of-00029.safetensors",
230
+ "layers.25.self_attn.o_proj.weight": "model-00012-of-00029.safetensors",
231
+ "layers.25.self_attn.q_proj.bias": "model-00012-of-00029.safetensors",
232
+ "layers.25.self_attn.q_proj.weight": "model-00012-of-00029.safetensors",
233
+ "layers.25.self_attn.v_proj.bias": "model-00012-of-00029.safetensors",
234
+ "layers.25.self_attn.v_proj.weight": "model-00012-of-00029.safetensors",
235
+ "layers.26.input_layernorm.weight": "model-00013-of-00029.safetensors",
236
+ "layers.26.mlp.down_proj.weight": "model-00013-of-00029.safetensors",
237
+ "layers.26.mlp.gate_proj.weight": "model-00012-of-00029.safetensors",
238
+ "layers.26.mlp.up_proj.weight": "model-00013-of-00029.safetensors",
239
+ "layers.26.post_attention_layernorm.weight": "model-00013-of-00029.safetensors",
240
+ "layers.26.self_attn.k_proj.bias": "model-00012-of-00029.safetensors",
241
+ "layers.26.self_attn.k_proj.weight": "model-00012-of-00029.safetensors",
242
+ "layers.26.self_attn.o_proj.weight": "model-00012-of-00029.safetensors",
243
+ "layers.26.self_attn.q_proj.bias": "model-00012-of-00029.safetensors",
244
+ "layers.26.self_attn.q_proj.weight": "model-00012-of-00029.safetensors",
245
+ "layers.26.self_attn.v_proj.bias": "model-00012-of-00029.safetensors",
246
+ "layers.26.self_attn.v_proj.weight": "model-00012-of-00029.safetensors",
247
+ "layers.27.input_layernorm.weight": "model-00013-of-00029.safetensors",
248
+ "layers.27.mlp.down_proj.weight": "model-00013-of-00029.safetensors",
249
+ "layers.27.mlp.gate_proj.weight": "model-00013-of-00029.safetensors",
250
+ "layers.27.mlp.up_proj.weight": "model-00013-of-00029.safetensors",
251
+ "layers.27.post_attention_layernorm.weight": "model-00013-of-00029.safetensors",
252
+ "layers.27.self_attn.k_proj.bias": "model-00013-of-00029.safetensors",
253
+ "layers.27.self_attn.k_proj.weight": "model-00013-of-00029.safetensors",
254
+ "layers.27.self_attn.o_proj.weight": "model-00013-of-00029.safetensors",
255
+ "layers.27.self_attn.q_proj.bias": "model-00013-of-00029.safetensors",
256
+ "layers.27.self_attn.q_proj.weight": "model-00013-of-00029.safetensors",
257
+ "layers.27.self_attn.v_proj.bias": "model-00013-of-00029.safetensors",
258
+ "layers.27.self_attn.v_proj.weight": "model-00013-of-00029.safetensors",
259
+ "layers.28.input_layernorm.weight": "model-00014-of-00029.safetensors",
260
+ "layers.28.mlp.down_proj.weight": "model-00014-of-00029.safetensors",
261
+ "layers.28.mlp.gate_proj.weight": "model-00013-of-00029.safetensors",
262
+ "layers.28.mlp.up_proj.weight": "model-00013-of-00029.safetensors",
263
+ "layers.28.post_attention_layernorm.weight": "model-00014-of-00029.safetensors",
264
+ "layers.28.self_attn.k_proj.bias": "model-00013-of-00029.safetensors",
265
+ "layers.28.self_attn.k_proj.weight": "model-00013-of-00029.safetensors",
266
+ "layers.28.self_attn.o_proj.weight": "model-00013-of-00029.safetensors",
267
+ "layers.28.self_attn.q_proj.bias": "model-00013-of-00029.safetensors",
268
+ "layers.28.self_attn.q_proj.weight": "model-00013-of-00029.safetensors",
269
+ "layers.28.self_attn.v_proj.bias": "model-00013-of-00029.safetensors",
270
+ "layers.28.self_attn.v_proj.weight": "model-00013-of-00029.safetensors",
271
+ "layers.29.input_layernorm.weight": "model-00014-of-00029.safetensors",
272
+ "layers.29.mlp.down_proj.weight": "model-00014-of-00029.safetensors",
273
+ "layers.29.mlp.gate_proj.weight": "model-00014-of-00029.safetensors",
274
+ "layers.29.mlp.up_proj.weight": "model-00014-of-00029.safetensors",
275
+ "layers.29.post_attention_layernorm.weight": "model-00014-of-00029.safetensors",
276
+ "layers.29.self_attn.k_proj.bias": "model-00014-of-00029.safetensors",
277
+ "layers.29.self_attn.k_proj.weight": "model-00014-of-00029.safetensors",
278
+ "layers.29.self_attn.o_proj.weight": "model-00014-of-00029.safetensors",
279
+ "layers.29.self_attn.q_proj.bias": "model-00014-of-00029.safetensors",
280
+ "layers.29.self_attn.q_proj.weight": "model-00014-of-00029.safetensors",
281
+ "layers.29.self_attn.v_proj.bias": "model-00014-of-00029.safetensors",
282
+ "layers.29.self_attn.v_proj.weight": "model-00014-of-00029.safetensors",
283
+ "layers.3.input_layernorm.weight": "model-00003-of-00029.safetensors",
284
+ "layers.3.mlp.down_proj.weight": "model-00003-of-00029.safetensors",
285
+ "layers.3.mlp.gate_proj.weight": "model-00003-of-00029.safetensors",
286
+ "layers.3.mlp.up_proj.weight": "model-00003-of-00029.safetensors",
287
+ "layers.3.post_attention_layernorm.weight": "model-00003-of-00029.safetensors",
288
+ "layers.3.self_attn.k_proj.bias": "model-00002-of-00029.safetensors",
289
+ "layers.3.self_attn.k_proj.weight": "model-00002-of-00029.safetensors",
290
+ "layers.3.self_attn.o_proj.weight": "model-00002-of-00029.safetensors",
291
+ "layers.3.self_attn.q_proj.bias": "model-00002-of-00029.safetensors",
292
+ "layers.3.self_attn.q_proj.weight": "model-00002-of-00029.safetensors",
293
+ "layers.3.self_attn.v_proj.bias": "model-00002-of-00029.safetensors",
294
+ "layers.3.self_attn.v_proj.weight": "model-00002-of-00029.safetensors",
295
+ "layers.30.input_layernorm.weight": "model-00014-of-00029.safetensors",
296
+ "layers.30.mlp.down_proj.weight": "model-00014-of-00029.safetensors",
297
+ "layers.30.mlp.gate_proj.weight": "model-00014-of-00029.safetensors",
298
+ "layers.30.mlp.up_proj.weight": "model-00014-of-00029.safetensors",
299
+ "layers.30.post_attention_layernorm.weight": "model-00014-of-00029.safetensors",
300
+ "layers.30.self_attn.k_proj.bias": "model-00014-of-00029.safetensors",
301
+ "layers.30.self_attn.k_proj.weight": "model-00014-of-00029.safetensors",
302
+ "layers.30.self_attn.o_proj.weight": "model-00014-of-00029.safetensors",
303
+ "layers.30.self_attn.q_proj.bias": "model-00014-of-00029.safetensors",
304
+ "layers.30.self_attn.q_proj.weight": "model-00014-of-00029.safetensors",
305
+ "layers.30.self_attn.v_proj.bias": "model-00014-of-00029.safetensors",
306
+ "layers.30.self_attn.v_proj.weight": "model-00014-of-00029.safetensors",
307
+ "layers.31.input_layernorm.weight": "model-00015-of-00029.safetensors",
308
+ "layers.31.mlp.down_proj.weight": "model-00015-of-00029.safetensors",
309
+ "layers.31.mlp.gate_proj.weight": "model-00015-of-00029.safetensors",
310
+ "layers.31.mlp.up_proj.weight": "model-00015-of-00029.safetensors",
311
+ "layers.31.post_attention_layernorm.weight": "model-00015-of-00029.safetensors",
312
+ "layers.31.self_attn.k_proj.bias": "model-00014-of-00029.safetensors",
313
+ "layers.31.self_attn.k_proj.weight": "model-00014-of-00029.safetensors",
314
+ "layers.31.self_attn.o_proj.weight": "model-00014-of-00029.safetensors",
315
+ "layers.31.self_attn.q_proj.bias": "model-00014-of-00029.safetensors",
316
+ "layers.31.self_attn.q_proj.weight": "model-00014-of-00029.safetensors",
317
+ "layers.31.self_attn.v_proj.bias": "model-00014-of-00029.safetensors",
318
+ "layers.31.self_attn.v_proj.weight": "model-00014-of-00029.safetensors",
319
+ "layers.32.input_layernorm.weight": "model-00015-of-00029.safetensors",
320
+ "layers.32.mlp.down_proj.weight": "model-00015-of-00029.safetensors",
321
+ "layers.32.mlp.gate_proj.weight": "model-00015-of-00029.safetensors",
322
+ "layers.32.mlp.up_proj.weight": "model-00015-of-00029.safetensors",
323
+ "layers.32.post_attention_layernorm.weight": "model-00015-of-00029.safetensors",
324
+ "layers.32.self_attn.k_proj.bias": "model-00015-of-00029.safetensors",
325
+ "layers.32.self_attn.k_proj.weight": "model-00015-of-00029.safetensors",
326
+ "layers.32.self_attn.o_proj.weight": "model-00015-of-00029.safetensors",
327
+ "layers.32.self_attn.q_proj.bias": "model-00015-of-00029.safetensors",
328
+ "layers.32.self_attn.q_proj.weight": "model-00015-of-00029.safetensors",
329
+ "layers.32.self_attn.v_proj.bias": "model-00015-of-00029.safetensors",
330
+ "layers.32.self_attn.v_proj.weight": "model-00015-of-00029.safetensors",
331
+ "layers.33.input_layernorm.weight": "model-00016-of-00029.safetensors",
332
+ "layers.33.mlp.down_proj.weight": "model-00016-of-00029.safetensors",
333
+ "layers.33.mlp.gate_proj.weight": "model-00015-of-00029.safetensors",
334
+ "layers.33.mlp.up_proj.weight": "model-00016-of-00029.safetensors",
335
+ "layers.33.post_attention_layernorm.weight": "model-00016-of-00029.safetensors",
336
+ "layers.33.self_attn.k_proj.bias": "model-00015-of-00029.safetensors",
337
+ "layers.33.self_attn.k_proj.weight": "model-00015-of-00029.safetensors",
338
+ "layers.33.self_attn.o_proj.weight": "model-00015-of-00029.safetensors",
339
+ "layers.33.self_attn.q_proj.bias": "model-00015-of-00029.safetensors",
340
+ "layers.33.self_attn.q_proj.weight": "model-00015-of-00029.safetensors",
341
+ "layers.33.self_attn.v_proj.bias": "model-00015-of-00029.safetensors",
342
+ "layers.33.self_attn.v_proj.weight": "model-00015-of-00029.safetensors",
343
+ "layers.34.input_layernorm.weight": "model-00016-of-00029.safetensors",
344
+ "layers.34.mlp.down_proj.weight": "model-00016-of-00029.safetensors",
345
+ "layers.34.mlp.gate_proj.weight": "model-00016-of-00029.safetensors",
346
+ "layers.34.mlp.up_proj.weight": "model-00016-of-00029.safetensors",
347
+ "layers.34.post_attention_layernorm.weight": "model-00016-of-00029.safetensors",
348
+ "layers.34.self_attn.k_proj.bias": "model-00016-of-00029.safetensors",
349
+ "layers.34.self_attn.k_proj.weight": "model-00016-of-00029.safetensors",
350
+ "layers.34.self_attn.o_proj.weight": "model-00016-of-00029.safetensors",
351
+ "layers.34.self_attn.q_proj.bias": "model-00016-of-00029.safetensors",
352
+ "layers.34.self_attn.q_proj.weight": "model-00016-of-00029.safetensors",
353
+ "layers.34.self_attn.v_proj.bias": "model-00016-of-00029.safetensors",
354
+ "layers.34.self_attn.v_proj.weight": "model-00016-of-00029.safetensors",
355
+ "layers.35.input_layernorm.weight": "model-00017-of-00029.safetensors",
356
+ "layers.35.mlp.down_proj.weight": "model-00017-of-00029.safetensors",
357
+ "layers.35.mlp.gate_proj.weight": "model-00016-of-00029.safetensors",
358
+ "layers.35.mlp.up_proj.weight": "model-00016-of-00029.safetensors",
359
+ "layers.35.post_attention_layernorm.weight": "model-00017-of-00029.safetensors",
360
+ "layers.35.self_attn.k_proj.bias": "model-00016-of-00029.safetensors",
361
+ "layers.35.self_attn.k_proj.weight": "model-00016-of-00029.safetensors",
362
+ "layers.35.self_attn.o_proj.weight": "model-00016-of-00029.safetensors",
363
+ "layers.35.self_attn.q_proj.bias": "model-00016-of-00029.safetensors",
364
+ "layers.35.self_attn.q_proj.weight": "model-00016-of-00029.safetensors",
365
+ "layers.35.self_attn.v_proj.bias": "model-00016-of-00029.safetensors",
366
+ "layers.35.self_attn.v_proj.weight": "model-00016-of-00029.safetensors",
367
+ "layers.36.input_layernorm.weight": "model-00017-of-00029.safetensors",
368
+ "layers.36.mlp.down_proj.weight": "model-00017-of-00029.safetensors",
369
+ "layers.36.mlp.gate_proj.weight": "model-00017-of-00029.safetensors",
370
+ "layers.36.mlp.up_proj.weight": "model-00017-of-00029.safetensors",
371
+ "layers.36.post_attention_layernorm.weight": "model-00017-of-00029.safetensors",
372
+ "layers.36.self_attn.k_proj.bias": "model-00017-of-00029.safetensors",
373
+ "layers.36.self_attn.k_proj.weight": "model-00017-of-00029.safetensors",
374
+ "layers.36.self_attn.o_proj.weight": "model-00017-of-00029.safetensors",
375
+ "layers.36.self_attn.q_proj.bias": "model-00017-of-00029.safetensors",
376
+ "layers.36.self_attn.q_proj.weight": "model-00017-of-00029.safetensors",
377
+ "layers.36.self_attn.v_proj.bias": "model-00017-of-00029.safetensors",
378
+ "layers.36.self_attn.v_proj.weight": "model-00017-of-00029.safetensors",
379
+ "layers.37.input_layernorm.weight": "model-00017-of-00029.safetensors",
380
+ "layers.37.mlp.down_proj.weight": "model-00017-of-00029.safetensors",
381
+ "layers.37.mlp.gate_proj.weight": "model-00017-of-00029.safetensors",
382
+ "layers.37.mlp.up_proj.weight": "model-00017-of-00029.safetensors",
383
+ "layers.37.post_attention_layernorm.weight": "model-00017-of-00029.safetensors",
384
+ "layers.37.self_attn.k_proj.bias": "model-00017-of-00029.safetensors",
385
+ "layers.37.self_attn.k_proj.weight": "model-00017-of-00029.safetensors",
386
+ "layers.37.self_attn.o_proj.weight": "model-00017-of-00029.safetensors",
387
+ "layers.37.self_attn.q_proj.bias": "model-00017-of-00029.safetensors",
388
+ "layers.37.self_attn.q_proj.weight": "model-00017-of-00029.safetensors",
389
+ "layers.37.self_attn.v_proj.bias": "model-00017-of-00029.safetensors",
390
+ "layers.37.self_attn.v_proj.weight": "model-00017-of-00029.safetensors",
391
+ "layers.38.input_layernorm.weight": "model-00018-of-00029.safetensors",
392
+ "layers.38.mlp.down_proj.weight": "model-00018-of-00029.safetensors",
393
+ "layers.38.mlp.gate_proj.weight": "model-00018-of-00029.safetensors",
394
+ "layers.38.mlp.up_proj.weight": "model-00018-of-00029.safetensors",
395
+ "layers.38.post_attention_layernorm.weight": "model-00018-of-00029.safetensors",
396
+ "layers.38.self_attn.k_proj.bias": "model-00017-of-00029.safetensors",
397
+ "layers.38.self_attn.k_proj.weight": "model-00017-of-00029.safetensors",
398
+ "layers.38.self_attn.o_proj.weight": "model-00017-of-00029.safetensors",
399
+ "layers.38.self_attn.q_proj.bias": "model-00017-of-00029.safetensors",
400
+ "layers.38.self_attn.q_proj.weight": "model-00017-of-00029.safetensors",
401
+ "layers.38.self_attn.v_proj.bias": "model-00017-of-00029.safetensors",
402
+ "layers.38.self_attn.v_proj.weight": "model-00017-of-00029.safetensors",
403
+ "layers.39.input_layernorm.weight": "model-00018-of-00029.safetensors",
404
+ "layers.39.mlp.down_proj.weight": "model-00018-of-00029.safetensors",
405
+ "layers.39.mlp.gate_proj.weight": "model-00018-of-00029.safetensors",
406
+ "layers.39.mlp.up_proj.weight": "model-00018-of-00029.safetensors",
407
+ "layers.39.post_attention_layernorm.weight": "model-00018-of-00029.safetensors",
408
+ "layers.39.self_attn.k_proj.bias": "model-00018-of-00029.safetensors",
409
+ "layers.39.self_attn.k_proj.weight": "model-00018-of-00029.safetensors",
410
+ "layers.39.self_attn.o_proj.weight": "model-00018-of-00029.safetensors",
411
+ "layers.39.self_attn.q_proj.bias": "model-00018-of-00029.safetensors",
412
+ "layers.39.self_attn.q_proj.weight": "model-00018-of-00029.safetensors",
413
+ "layers.39.self_attn.v_proj.bias": "model-00018-of-00029.safetensors",
414
+ "layers.39.self_attn.v_proj.weight": "model-00018-of-00029.safetensors",
415
+ "layers.4.input_layernorm.weight": "model-00003-of-00029.safetensors",
416
+ "layers.4.mlp.down_proj.weight": "model-00003-of-00029.safetensors",
417
+ "layers.4.mlp.gate_proj.weight": "model-00003-of-00029.safetensors",
418
+ "layers.4.mlp.up_proj.weight": "model-00003-of-00029.safetensors",
419
+ "layers.4.post_attention_layernorm.weight": "model-00003-of-00029.safetensors",
420
+ "layers.4.self_attn.k_proj.bias": "model-00003-of-00029.safetensors",
421
+ "layers.4.self_attn.k_proj.weight": "model-00003-of-00029.safetensors",
422
+ "layers.4.self_attn.o_proj.weight": "model-00003-of-00029.safetensors",
423
+ "layers.4.self_attn.q_proj.bias": "model-00003-of-00029.safetensors",
424
+ "layers.4.self_attn.q_proj.weight": "model-00003-of-00029.safetensors",
425
+ "layers.4.self_attn.v_proj.bias": "model-00003-of-00029.safetensors",
426
+ "layers.4.self_attn.v_proj.weight": "model-00003-of-00029.safetensors",
427
+ "layers.40.input_layernorm.weight": "model-00019-of-00029.safetensors",
428
+ "layers.40.mlp.down_proj.weight": "model-00019-of-00029.safetensors",
429
+ "layers.40.mlp.gate_proj.weight": "model-00018-of-00029.safetensors",
430
+ "layers.40.mlp.up_proj.weight": "model-00019-of-00029.safetensors",
431
+ "layers.40.post_attention_layernorm.weight": "model-00019-of-00029.safetensors",
432
+ "layers.40.self_attn.k_proj.bias": "model-00018-of-00029.safetensors",
433
+ "layers.40.self_attn.k_proj.weight": "model-00018-of-00029.safetensors",
434
+ "layers.40.self_attn.o_proj.weight": "model-00018-of-00029.safetensors",
435
+ "layers.40.self_attn.q_proj.bias": "model-00018-of-00029.safetensors",
436
+ "layers.40.self_attn.q_proj.weight": "model-00018-of-00029.safetensors",
437
+ "layers.40.self_attn.v_proj.bias": "model-00018-of-00029.safetensors",
438
+ "layers.40.self_attn.v_proj.weight": "model-00018-of-00029.safetensors",
439
+ "layers.41.input_layernorm.weight": "model-00019-of-00029.safetensors",
440
+ "layers.41.mlp.down_proj.weight": "model-00019-of-00029.safetensors",
441
+ "layers.41.mlp.gate_proj.weight": "model-00019-of-00029.safetensors",
442
+ "layers.41.mlp.up_proj.weight": "model-00019-of-00029.safetensors",
443
+ "layers.41.post_attention_layernorm.weight": "model-00019-of-00029.safetensors",
444
+ "layers.41.self_attn.k_proj.bias": "model-00019-of-00029.safetensors",
445
+ "layers.41.self_attn.k_proj.weight": "model-00019-of-00029.safetensors",
446
+ "layers.41.self_attn.o_proj.weight": "model-00019-of-00029.safetensors",
447
+ "layers.41.self_attn.q_proj.bias": "model-00019-of-00029.safetensors",
448
+ "layers.41.self_attn.q_proj.weight": "model-00019-of-00029.safetensors",
449
+ "layers.41.self_attn.v_proj.bias": "model-00019-of-00029.safetensors",
450
+ "layers.41.self_attn.v_proj.weight": "model-00019-of-00029.safetensors",
451
+ "layers.42.input_layernorm.weight": "model-00020-of-00029.safetensors",
452
+ "layers.42.mlp.down_proj.weight": "model-00020-of-00029.safetensors",
453
+ "layers.42.mlp.gate_proj.weight": "model-00019-of-00029.safetensors",
454
+ "layers.42.mlp.up_proj.weight": "model-00019-of-00029.safetensors",
455
+ "layers.42.post_attention_layernorm.weight": "model-00020-of-00029.safetensors",
456
+ "layers.42.self_attn.k_proj.bias": "model-00019-of-00029.safetensors",
457
+ "layers.42.self_attn.k_proj.weight": "model-00019-of-00029.safetensors",
458
+ "layers.42.self_attn.o_proj.weight": "model-00019-of-00029.safetensors",
459
+ "layers.42.self_attn.q_proj.bias": "model-00019-of-00029.safetensors",
460
+ "layers.42.self_attn.q_proj.weight": "model-00019-of-00029.safetensors",
461
+ "layers.42.self_attn.v_proj.bias": "model-00019-of-00029.safetensors",
462
+ "layers.42.self_attn.v_proj.weight": "model-00019-of-00029.safetensors",
463
+ "layers.43.input_layernorm.weight": "model-00020-of-00029.safetensors",
464
+ "layers.43.mlp.down_proj.weight": "model-00020-of-00029.safetensors",
465
+ "layers.43.mlp.gate_proj.weight": "model-00020-of-00029.safetensors",
466
+ "layers.43.mlp.up_proj.weight": "model-00020-of-00029.safetensors",
467
+ "layers.43.post_attention_layernorm.weight": "model-00020-of-00029.safetensors",
468
+ "layers.43.self_attn.k_proj.bias": "model-00020-of-00029.safetensors",
469
+ "layers.43.self_attn.k_proj.weight": "model-00020-of-00029.safetensors",
470
+ "layers.43.self_attn.o_proj.weight": "model-00020-of-00029.safetensors",
471
+ "layers.43.self_attn.q_proj.bias": "model-00020-of-00029.safetensors",
472
+ "layers.43.self_attn.q_proj.weight": "model-00020-of-00029.safetensors",
473
+ "layers.43.self_attn.v_proj.bias": "model-00020-of-00029.safetensors",
474
+ "layers.43.self_attn.v_proj.weight": "model-00020-of-00029.safetensors",
475
+ "layers.44.input_layernorm.weight": "model-00020-of-00029.safetensors",
476
+ "layers.44.mlp.down_proj.weight": "model-00020-of-00029.safetensors",
477
+ "layers.44.mlp.gate_proj.weight": "model-00020-of-00029.safetensors",
478
+ "layers.44.mlp.up_proj.weight": "model-00020-of-00029.safetensors",
479
+ "layers.44.post_attention_layernorm.weight": "model-00020-of-00029.safetensors",
480
+ "layers.44.self_attn.k_proj.bias": "model-00020-of-00029.safetensors",
481
+ "layers.44.self_attn.k_proj.weight": "model-00020-of-00029.safetensors",
482
+ "layers.44.self_attn.o_proj.weight": "model-00020-of-00029.safetensors",
483
+ "layers.44.self_attn.q_proj.bias": "model-00020-of-00029.safetensors",
484
+ "layers.44.self_attn.q_proj.weight": "model-00020-of-00029.safetensors",
485
+ "layers.44.self_attn.v_proj.bias": "model-00020-of-00029.safetensors",
486
+ "layers.44.self_attn.v_proj.weight": "model-00020-of-00029.safetensors",
487
+ "layers.45.input_layernorm.weight": "model-00021-of-00029.safetensors",
488
+ "layers.45.mlp.down_proj.weight": "model-00021-of-00029.safetensors",
489
+ "layers.45.mlp.gate_proj.weight": "model-00021-of-00029.safetensors",
490
+ "layers.45.mlp.up_proj.weight": "model-00021-of-00029.safetensors",
491
+ "layers.45.post_attention_layernorm.weight": "model-00021-of-00029.safetensors",
492
+ "layers.45.self_attn.k_proj.bias": "model-00020-of-00029.safetensors",
493
+ "layers.45.self_attn.k_proj.weight": "model-00020-of-00029.safetensors",
494
+ "layers.45.self_attn.o_proj.weight": "model-00020-of-00029.safetensors",
495
+ "layers.45.self_attn.q_proj.bias": "model-00020-of-00029.safetensors",
496
+ "layers.45.self_attn.q_proj.weight": "model-00020-of-00029.safetensors",
497
+ "layers.45.self_attn.v_proj.bias": "model-00020-of-00029.safetensors",
498
+ "layers.45.self_attn.v_proj.weight": "model-00020-of-00029.safetensors",
499
+ "layers.46.input_layernorm.weight": "model-00021-of-00029.safetensors",
500
+ "layers.46.mlp.down_proj.weight": "model-00021-of-00029.safetensors",
501
+ "layers.46.mlp.gate_proj.weight": "model-00021-of-00029.safetensors",
502
+ "layers.46.mlp.up_proj.weight": "model-00021-of-00029.safetensors",
503
+ "layers.46.post_attention_layernorm.weight": "model-00021-of-00029.safetensors",
504
+ "layers.46.self_attn.k_proj.bias": "model-00021-of-00029.safetensors",
505
+ "layers.46.self_attn.k_proj.weight": "model-00021-of-00029.safetensors",
506
+ "layers.46.self_attn.o_proj.weight": "model-00021-of-00029.safetensors",
507
+ "layers.46.self_attn.q_proj.bias": "model-00021-of-00029.safetensors",
508
+ "layers.46.self_attn.q_proj.weight": "model-00021-of-00029.safetensors",
509
+ "layers.46.self_attn.v_proj.bias": "model-00021-of-00029.safetensors",
510
+ "layers.46.self_attn.v_proj.weight": "model-00021-of-00029.safetensors",
511
+ "layers.47.input_layernorm.weight": "model-00022-of-00029.safetensors",
512
+ "layers.47.mlp.down_proj.weight": "model-00022-of-00029.safetensors",
513
+ "layers.47.mlp.gate_proj.weight": "model-00021-of-00029.safetensors",
514
+ "layers.47.mlp.up_proj.weight": "model-00022-of-00029.safetensors",
515
+ "layers.47.post_attention_layernorm.weight": "model-00022-of-00029.safetensors",
516
+ "layers.47.self_attn.k_proj.bias": "model-00021-of-00029.safetensors",
517
+ "layers.47.self_attn.k_proj.weight": "model-00021-of-00029.safetensors",
518
+ "layers.47.self_attn.o_proj.weight": "model-00021-of-00029.safetensors",
519
+ "layers.47.self_attn.q_proj.bias": "model-00021-of-00029.safetensors",
520
+ "layers.47.self_attn.q_proj.weight": "model-00021-of-00029.safetensors",
521
+ "layers.47.self_attn.v_proj.bias": "model-00021-of-00029.safetensors",
522
+ "layers.47.self_attn.v_proj.weight": "model-00021-of-00029.safetensors",
523
+ "layers.48.input_layernorm.weight": "model-00022-of-00029.safetensors",
524
+ "layers.48.mlp.down_proj.weight": "model-00022-of-00029.safetensors",
525
+ "layers.48.mlp.gate_proj.weight": "model-00022-of-00029.safetensors",
526
+ "layers.48.mlp.up_proj.weight": "model-00022-of-00029.safetensors",
527
+ "layers.48.post_attention_layernorm.weight": "model-00022-of-00029.safetensors",
528
+ "layers.48.self_attn.k_proj.bias": "model-00022-of-00029.safetensors",
529
+ "layers.48.self_attn.k_proj.weight": "model-00022-of-00029.safetensors",
530
+ "layers.48.self_attn.o_proj.weight": "model-00022-of-00029.safetensors",
531
+ "layers.48.self_attn.q_proj.bias": "model-00022-of-00029.safetensors",
532
+ "layers.48.self_attn.q_proj.weight": "model-00022-of-00029.safetensors",
533
+ "layers.48.self_attn.v_proj.bias": "model-00022-of-00029.safetensors",
534
+ "layers.48.self_attn.v_proj.weight": "model-00022-of-00029.safetensors",
535
+ "layers.49.input_layernorm.weight": "model-00023-of-00029.safetensors",
536
+ "layers.49.mlp.down_proj.weight": "model-00023-of-00029.safetensors",
537
+ "layers.49.mlp.gate_proj.weight": "model-00022-of-00029.safetensors",
538
+ "layers.49.mlp.up_proj.weight": "model-00022-of-00029.safetensors",
539
+ "layers.49.post_attention_layernorm.weight": "model-00023-of-00029.safetensors",
540
+ "layers.49.self_attn.k_proj.bias": "model-00022-of-00029.safetensors",
541
+ "layers.49.self_attn.k_proj.weight": "model-00022-of-00029.safetensors",
542
+ "layers.49.self_attn.o_proj.weight": "model-00022-of-00029.safetensors",
543
+ "layers.49.self_attn.q_proj.bias": "model-00022-of-00029.safetensors",
544
+ "layers.49.self_attn.q_proj.weight": "model-00022-of-00029.safetensors",
545
+ "layers.49.self_attn.v_proj.bias": "model-00022-of-00029.safetensors",
546
+ "layers.49.self_attn.v_proj.weight": "model-00022-of-00029.safetensors",
547
+ "layers.5.input_layernorm.weight": "model-00004-of-00029.safetensors",
548
+ "layers.5.mlp.down_proj.weight": "model-00004-of-00029.safetensors",
549
+ "layers.5.mlp.gate_proj.weight": "model-00003-of-00029.safetensors",
550
+ "layers.5.mlp.up_proj.weight": "model-00004-of-00029.safetensors",
551
+ "layers.5.post_attention_layernorm.weight": "model-00004-of-00029.safetensors",
552
+ "layers.5.self_attn.k_proj.bias": "model-00003-of-00029.safetensors",
553
+ "layers.5.self_attn.k_proj.weight": "model-00003-of-00029.safetensors",
554
+ "layers.5.self_attn.o_proj.weight": "model-00003-of-00029.safetensors",
555
+ "layers.5.self_attn.q_proj.bias": "model-00003-of-00029.safetensors",
556
+ "layers.5.self_attn.q_proj.weight": "model-00003-of-00029.safetensors",
557
+ "layers.5.self_attn.v_proj.bias": "model-00003-of-00029.safetensors",
558
+ "layers.5.self_attn.v_proj.weight": "model-00003-of-00029.safetensors",
559
+ "layers.50.input_layernorm.weight": "model-00023-of-00029.safetensors",
560
+ "layers.50.mlp.down_proj.weight": "model-00023-of-00029.safetensors",
561
+ "layers.50.mlp.gate_proj.weight": "model-00023-of-00029.safetensors",
562
+ "layers.50.mlp.up_proj.weight": "model-00023-of-00029.safetensors",
563
+ "layers.50.post_attention_layernorm.weight": "model-00023-of-00029.safetensors",
564
+ "layers.50.self_attn.k_proj.bias": "model-00023-of-00029.safetensors",
565
+ "layers.50.self_attn.k_proj.weight": "model-00023-of-00029.safetensors",
566
+ "layers.50.self_attn.o_proj.weight": "model-00023-of-00029.safetensors",
567
+ "layers.50.self_attn.q_proj.bias": "model-00023-of-00029.safetensors",
568
+ "layers.50.self_attn.q_proj.weight": "model-00023-of-00029.safetensors",
569
+ "layers.50.self_attn.v_proj.bias": "model-00023-of-00029.safetensors",
570
+ "layers.50.self_attn.v_proj.weight": "model-00023-of-00029.safetensors",
571
+ "layers.51.input_layernorm.weight": "model-00023-of-00029.safetensors",
572
+ "layers.51.mlp.down_proj.weight": "model-00023-of-00029.safetensors",
573
+ "layers.51.mlp.gate_proj.weight": "model-00023-of-00029.safetensors",
574
+ "layers.51.mlp.up_proj.weight": "model-00023-of-00029.safetensors",
575
+ "layers.51.post_attention_layernorm.weight": "model-00023-of-00029.safetensors",
576
+ "layers.51.self_attn.k_proj.bias": "model-00023-of-00029.safetensors",
577
+ "layers.51.self_attn.k_proj.weight": "model-00023-of-00029.safetensors",
578
+ "layers.51.self_attn.o_proj.weight": "model-00023-of-00029.safetensors",
579
+ "layers.51.self_attn.q_proj.bias": "model-00023-of-00029.safetensors",
580
+ "layers.51.self_attn.q_proj.weight": "model-00023-of-00029.safetensors",
581
+ "layers.51.self_attn.v_proj.bias": "model-00023-of-00029.safetensors",
582
+ "layers.51.self_attn.v_proj.weight": "model-00023-of-00029.safetensors",
583
+ "layers.52.input_layernorm.weight": "model-00024-of-00029.safetensors",
584
+ "layers.52.mlp.down_proj.weight": "model-00024-of-00029.safetensors",
585
+ "layers.52.mlp.gate_proj.weight": "model-00024-of-00029.safetensors",
586
+ "layers.52.mlp.up_proj.weight": "model-00024-of-00029.safetensors",
587
+ "layers.52.post_attention_layernorm.weight": "model-00024-of-00029.safetensors",
588
+ "layers.52.self_attn.k_proj.bias": "model-00023-of-00029.safetensors",
589
+ "layers.52.self_attn.k_proj.weight": "model-00023-of-00029.safetensors",
590
+ "layers.52.self_attn.o_proj.weight": "model-00023-of-00029.safetensors",
591
+ "layers.52.self_attn.q_proj.bias": "model-00023-of-00029.safetensors",
592
+ "layers.52.self_attn.q_proj.weight": "model-00023-of-00029.safetensors",
593
+ "layers.52.self_attn.v_proj.bias": "model-00023-of-00029.safetensors",
594
+ "layers.52.self_attn.v_proj.weight": "model-00023-of-00029.safetensors",
595
+ "layers.53.input_layernorm.weight": "model-00024-of-00029.safetensors",
596
+ "layers.53.mlp.down_proj.weight": "model-00024-of-00029.safetensors",
597
+ "layers.53.mlp.gate_proj.weight": "model-00024-of-00029.safetensors",
598
+ "layers.53.mlp.up_proj.weight": "model-00024-of-00029.safetensors",
599
+ "layers.53.post_attention_layernorm.weight": "model-00024-of-00029.safetensors",
600
+ "layers.53.self_attn.k_proj.bias": "model-00024-of-00029.safetensors",
601
+ "layers.53.self_attn.k_proj.weight": "model-00024-of-00029.safetensors",
602
+ "layers.53.self_attn.o_proj.weight": "model-00024-of-00029.safetensors",
603
+ "layers.53.self_attn.q_proj.bias": "model-00024-of-00029.safetensors",
604
+ "layers.53.self_attn.q_proj.weight": "model-00024-of-00029.safetensors",
605
+ "layers.53.self_attn.v_proj.bias": "model-00024-of-00029.safetensors",
606
+ "layers.53.self_attn.v_proj.weight": "model-00024-of-00029.safetensors",
607
+ "layers.54.input_layernorm.weight": "model-00025-of-00029.safetensors",
608
+ "layers.54.mlp.down_proj.weight": "model-00025-of-00029.safetensors",
609
+ "layers.54.mlp.gate_proj.weight": "model-00024-of-00029.safetensors",
610
+ "layers.54.mlp.up_proj.weight": "model-00025-of-00029.safetensors",
611
+ "layers.54.post_attention_layernorm.weight": "model-00025-of-00029.safetensors",
612
+ "layers.54.self_attn.k_proj.bias": "model-00024-of-00029.safetensors",
613
+ "layers.54.self_attn.k_proj.weight": "model-00024-of-00029.safetensors",
614
+ "layers.54.self_attn.o_proj.weight": "model-00024-of-00029.safetensors",
615
+ "layers.54.self_attn.q_proj.bias": "model-00024-of-00029.safetensors",
616
+ "layers.54.self_attn.q_proj.weight": "model-00024-of-00029.safetensors",
617
+ "layers.54.self_attn.v_proj.bias": "model-00024-of-00029.safetensors",
618
+ "layers.54.self_attn.v_proj.weight": "model-00024-of-00029.safetensors",
619
+ "layers.55.input_layernorm.weight": "model-00025-of-00029.safetensors",
620
+ "layers.55.mlp.down_proj.weight": "model-00025-of-00029.safetensors",
621
+ "layers.55.mlp.gate_proj.weight": "model-00025-of-00029.safetensors",
622
+ "layers.55.mlp.up_proj.weight": "model-00025-of-00029.safetensors",
623
+ "layers.55.post_attention_layernorm.weight": "model-00025-of-00029.safetensors",
624
+ "layers.55.self_attn.k_proj.bias": "model-00025-of-00029.safetensors",
625
+ "layers.55.self_attn.k_proj.weight": "model-00025-of-00029.safetensors",
626
+ "layers.55.self_attn.o_proj.weight": "model-00025-of-00029.safetensors",
627
+ "layers.55.self_attn.q_proj.bias": "model-00025-of-00029.safetensors",
628
+ "layers.55.self_attn.q_proj.weight": "model-00025-of-00029.safetensors",
629
+ "layers.55.self_attn.v_proj.bias": "model-00025-of-00029.safetensors",
630
+ "layers.55.self_attn.v_proj.weight": "model-00025-of-00029.safetensors",
631
+ "layers.56.input_layernorm.weight": "model-00026-of-00029.safetensors",
632
+ "layers.56.mlp.down_proj.weight": "model-00026-of-00029.safetensors",
633
+ "layers.56.mlp.gate_proj.weight": "model-00025-of-00029.safetensors",
634
+ "layers.56.mlp.up_proj.weight": "model-00025-of-00029.safetensors",
635
+ "layers.56.post_attention_layernorm.weight": "model-00026-of-00029.safetensors",
636
+ "layers.56.self_attn.k_proj.bias": "model-00025-of-00029.safetensors",
637
+ "layers.56.self_attn.k_proj.weight": "model-00025-of-00029.safetensors",
638
+ "layers.56.self_attn.o_proj.weight": "model-00025-of-00029.safetensors",
639
+ "layers.56.self_attn.q_proj.bias": "model-00025-of-00029.safetensors",
640
+ "layers.56.self_attn.q_proj.weight": "model-00025-of-00029.safetensors",
641
+ "layers.56.self_attn.v_proj.bias": "model-00025-of-00029.safetensors",
642
+ "layers.56.self_attn.v_proj.weight": "model-00025-of-00029.safetensors",
643
+ "layers.57.input_layernorm.weight": "model-00026-of-00029.safetensors",
644
+ "layers.57.mlp.down_proj.weight": "model-00026-of-00029.safetensors",
645
+ "layers.57.mlp.gate_proj.weight": "model-00026-of-00029.safetensors",
646
+ "layers.57.mlp.up_proj.weight": "model-00026-of-00029.safetensors",
647
+ "layers.57.post_attention_layernorm.weight": "model-00026-of-00029.safetensors",
648
+ "layers.57.self_attn.k_proj.bias": "model-00026-of-00029.safetensors",
649
+ "layers.57.self_attn.k_proj.weight": "model-00026-of-00029.safetensors",
650
+ "layers.57.self_attn.o_proj.weight": "model-00026-of-00029.safetensors",
651
+ "layers.57.self_attn.q_proj.bias": "model-00026-of-00029.safetensors",
652
+ "layers.57.self_attn.q_proj.weight": "model-00026-of-00029.safetensors",
653
+ "layers.57.self_attn.v_proj.bias": "model-00026-of-00029.safetensors",
654
+ "layers.57.self_attn.v_proj.weight": "model-00026-of-00029.safetensors",
655
+ "layers.58.input_layernorm.weight": "model-00026-of-00029.safetensors",
656
+ "layers.58.mlp.down_proj.weight": "model-00026-of-00029.safetensors",
657
+ "layers.58.mlp.gate_proj.weight": "model-00026-of-00029.safetensors",
658
+ "layers.58.mlp.up_proj.weight": "model-00026-of-00029.safetensors",
659
+ "layers.58.post_attention_layernorm.weight": "model-00026-of-00029.safetensors",
660
+ "layers.58.self_attn.k_proj.bias": "model-00026-of-00029.safetensors",
661
+ "layers.58.self_attn.k_proj.weight": "model-00026-of-00029.safetensors",
662
+ "layers.58.self_attn.o_proj.weight": "model-00026-of-00029.safetensors",
663
+ "layers.58.self_attn.q_proj.bias": "model-00026-of-00029.safetensors",
664
+ "layers.58.self_attn.q_proj.weight": "model-00026-of-00029.safetensors",
665
+ "layers.58.self_attn.v_proj.bias": "model-00026-of-00029.safetensors",
666
+ "layers.58.self_attn.v_proj.weight": "model-00026-of-00029.safetensors",
667
+ "layers.59.input_layernorm.weight": "model-00027-of-00029.safetensors",
668
+ "layers.59.mlp.down_proj.weight": "model-00027-of-00029.safetensors",
669
+ "layers.59.mlp.gate_proj.weight": "model-00027-of-00029.safetensors",
670
+ "layers.59.mlp.up_proj.weight": "model-00027-of-00029.safetensors",
671
+ "layers.59.post_attention_layernorm.weight": "model-00027-of-00029.safetensors",
672
+ "layers.59.self_attn.k_proj.bias": "model-00026-of-00029.safetensors",
673
+ "layers.59.self_attn.k_proj.weight": "model-00026-of-00029.safetensors",
674
+ "layers.59.self_attn.o_proj.weight": "model-00026-of-00029.safetensors",
675
+ "layers.59.self_attn.q_proj.bias": "model-00026-of-00029.safetensors",
676
+ "layers.59.self_attn.q_proj.weight": "model-00026-of-00029.safetensors",
677
+ "layers.59.self_attn.v_proj.bias": "model-00026-of-00029.safetensors",
678
+ "layers.59.self_attn.v_proj.weight": "model-00026-of-00029.safetensors",
679
+ "layers.6.input_layernorm.weight": "model-00004-of-00029.safetensors",
680
+ "layers.6.mlp.down_proj.weight": "model-00004-of-00029.safetensors",
681
+ "layers.6.mlp.gate_proj.weight": "model-00004-of-00029.safetensors",
682
+ "layers.6.mlp.up_proj.weight": "model-00004-of-00029.safetensors",
683
+ "layers.6.post_attention_layernorm.weight": "model-00004-of-00029.safetensors",
684
+ "layers.6.self_attn.k_proj.bias": "model-00004-of-00029.safetensors",
685
+ "layers.6.self_attn.k_proj.weight": "model-00004-of-00029.safetensors",
686
+ "layers.6.self_attn.o_proj.weight": "model-00004-of-00029.safetensors",
687
+ "layers.6.self_attn.q_proj.bias": "model-00004-of-00029.safetensors",
688
+ "layers.6.self_attn.q_proj.weight": "model-00004-of-00029.safetensors",
689
+ "layers.6.self_attn.v_proj.bias": "model-00004-of-00029.safetensors",
690
+ "layers.6.self_attn.v_proj.weight": "model-00004-of-00029.safetensors",
691
+ "layers.60.input_layernorm.weight": "model-00027-of-00029.safetensors",
692
+ "layers.60.mlp.down_proj.weight": "model-00027-of-00029.safetensors",
693
+ "layers.60.mlp.gate_proj.weight": "model-00027-of-00029.safetensors",
694
+ "layers.60.mlp.up_proj.weight": "model-00027-of-00029.safetensors",
695
+ "layers.60.post_attention_layernorm.weight": "model-00027-of-00029.safetensors",
696
+ "layers.60.self_attn.k_proj.bias": "model-00027-of-00029.safetensors",
697
+ "layers.60.self_attn.k_proj.weight": "model-00027-of-00029.safetensors",
698
+ "layers.60.self_attn.o_proj.weight": "model-00027-of-00029.safetensors",
699
+ "layers.60.self_attn.q_proj.bias": "model-00027-of-00029.safetensors",
700
+ "layers.60.self_attn.q_proj.weight": "model-00027-of-00029.safetensors",
701
+ "layers.60.self_attn.v_proj.bias": "model-00027-of-00029.safetensors",
702
+ "layers.60.self_attn.v_proj.weight": "model-00027-of-00029.safetensors",
703
+ "layers.61.input_layernorm.weight": "model-00028-of-00029.safetensors",
704
+ "layers.61.mlp.down_proj.weight": "model-00028-of-00029.safetensors",
705
+ "layers.61.mlp.gate_proj.weight": "model-00027-of-00029.safetensors",
706
+ "layers.61.mlp.up_proj.weight": "model-00028-of-00029.safetensors",
707
+ "layers.61.post_attention_layernorm.weight": "model-00028-of-00029.safetensors",
708
+ "layers.61.self_attn.k_proj.bias": "model-00027-of-00029.safetensors",
709
+ "layers.61.self_attn.k_proj.weight": "model-00027-of-00029.safetensors",
710
+ "layers.61.self_attn.o_proj.weight": "model-00027-of-00029.safetensors",
711
+ "layers.61.self_attn.q_proj.bias": "model-00027-of-00029.safetensors",
712
+ "layers.61.self_attn.q_proj.weight": "model-00027-of-00029.safetensors",
713
+ "layers.61.self_attn.v_proj.bias": "model-00027-of-00029.safetensors",
714
+ "layers.61.self_attn.v_proj.weight": "model-00027-of-00029.safetensors",
715
+ "layers.62.input_layernorm.weight": "model-00028-of-00029.safetensors",
716
+ "layers.62.mlp.down_proj.weight": "model-00028-of-00029.safetensors",
717
+ "layers.62.mlp.gate_proj.weight": "model-00028-of-00029.safetensors",
718
+ "layers.62.mlp.up_proj.weight": "model-00028-of-00029.safetensors",
719
+ "layers.62.post_attention_layernorm.weight": "model-00028-of-00029.safetensors",
720
+ "layers.62.self_attn.k_proj.bias": "model-00028-of-00029.safetensors",
721
+ "layers.62.self_attn.k_proj.weight": "model-00028-of-00029.safetensors",
722
+ "layers.62.self_attn.o_proj.weight": "model-00028-of-00029.safetensors",
723
+ "layers.62.self_attn.q_proj.bias": "model-00028-of-00029.safetensors",
724
+ "layers.62.self_attn.q_proj.weight": "model-00028-of-00029.safetensors",
725
+ "layers.62.self_attn.v_proj.bias": "model-00028-of-00029.safetensors",
726
+ "layers.62.self_attn.v_proj.weight": "model-00028-of-00029.safetensors",
727
+ "layers.63.input_layernorm.weight": "model-00029-of-00029.safetensors",
728
+ "layers.63.mlp.down_proj.weight": "model-00029-of-00029.safetensors",
729
+ "layers.63.mlp.gate_proj.weight": "model-00028-of-00029.safetensors",
730
+ "layers.63.mlp.up_proj.weight": "model-00028-of-00029.safetensors",
731
+ "layers.63.post_attention_layernorm.weight": "model-00029-of-00029.safetensors",
732
+ "layers.63.self_attn.k_proj.bias": "model-00028-of-00029.safetensors",
733
+ "layers.63.self_attn.k_proj.weight": "model-00028-of-00029.safetensors",
734
+ "layers.63.self_attn.o_proj.weight": "model-00028-of-00029.safetensors",
735
+ "layers.63.self_attn.q_proj.bias": "model-00028-of-00029.safetensors",
736
+ "layers.63.self_attn.q_proj.weight": "model-00028-of-00029.safetensors",
737
+ "layers.63.self_attn.v_proj.bias": "model-00028-of-00029.safetensors",
738
+ "layers.63.self_attn.v_proj.weight": "model-00028-of-00029.safetensors",
739
+ "layers.7.input_layernorm.weight": "model-00005-of-00029.safetensors",
740
+ "layers.7.mlp.down_proj.weight": "model-00005-of-00029.safetensors",
741
+ "layers.7.mlp.gate_proj.weight": "model-00004-of-00029.safetensors",
742
+ "layers.7.mlp.up_proj.weight": "model-00004-of-00029.safetensors",
743
+ "layers.7.post_attention_layernorm.weight": "model-00005-of-00029.safetensors",
744
+ "layers.7.self_attn.k_proj.bias": "model-00004-of-00029.safetensors",
745
+ "layers.7.self_attn.k_proj.weight": "model-00004-of-00029.safetensors",
746
+ "layers.7.self_attn.o_proj.weight": "model-00004-of-00029.safetensors",
747
+ "layers.7.self_attn.q_proj.bias": "model-00004-of-00029.safetensors",
748
+ "layers.7.self_attn.q_proj.weight": "model-00004-of-00029.safetensors",
749
+ "layers.7.self_attn.v_proj.bias": "model-00004-of-00029.safetensors",
750
+ "layers.7.self_attn.v_proj.weight": "model-00004-of-00029.safetensors",
751
+ "layers.8.input_layernorm.weight": "model-00005-of-00029.safetensors",
752
+ "layers.8.mlp.down_proj.weight": "model-00005-of-00029.safetensors",
753
+ "layers.8.mlp.gate_proj.weight": "model-00005-of-00029.safetensors",
754
+ "layers.8.mlp.up_proj.weight": "model-00005-of-00029.safetensors",
755
+ "layers.8.post_attention_layernorm.weight": "model-00005-of-00029.safetensors",
756
+ "layers.8.self_attn.k_proj.bias": "model-00005-of-00029.safetensors",
757
+ "layers.8.self_attn.k_proj.weight": "model-00005-of-00029.safetensors",
758
+ "layers.8.self_attn.o_proj.weight": "model-00005-of-00029.safetensors",
759
+ "layers.8.self_attn.q_proj.bias": "model-00005-of-00029.safetensors",
760
+ "layers.8.self_attn.q_proj.weight": "model-00005-of-00029.safetensors",
761
+ "layers.8.self_attn.v_proj.bias": "model-00005-of-00029.safetensors",
762
+ "layers.8.self_attn.v_proj.weight": "model-00005-of-00029.safetensors",
763
+ "layers.9.input_layernorm.weight": "model-00005-of-00029.safetensors",
764
+ "layers.9.mlp.down_proj.weight": "model-00005-of-00029.safetensors",
765
+ "layers.9.mlp.gate_proj.weight": "model-00005-of-00029.safetensors",
766
+ "layers.9.mlp.up_proj.weight": "model-00005-of-00029.safetensors",
767
+ "layers.9.post_attention_layernorm.weight": "model-00005-of-00029.safetensors",
768
+ "layers.9.self_attn.k_proj.bias": "model-00005-of-00029.safetensors",
769
+ "layers.9.self_attn.k_proj.weight": "model-00005-of-00029.safetensors",
770
+ "layers.9.self_attn.o_proj.weight": "model-00005-of-00029.safetensors",
771
+ "layers.9.self_attn.q_proj.bias": "model-00005-of-00029.safetensors",
772
+ "layers.9.self_attn.q_proj.weight": "model-00005-of-00029.safetensors",
773
+ "layers.9.self_attn.v_proj.bias": "model-00005-of-00029.safetensors",
774
+ "layers.9.self_attn.v_proj.weight": "model-00005-of-00029.safetensors",
775
+ "norm.weight": "model-00029-of-00029.safetensors"
776
+ }
777
+ }