Hugging Face's logo

Eiki
/
btx

Transformers
Safetensors
mixtral
text-generation-inference
Model card Files Community
Eiki commited on
Commit
f5ae327
·
verified ·
1 Parent(s): 817bd47

Upload folder using huggingface_hub

Browse files
Files changed (9) hide show
  1. config.json +35 -0
  2. generation_config.json +6 -0
  3. model-00001-of-00004.safetensors +3 -0
  4. model-00002-of-00004.safetensors +3 -0
  5. model-00003-of-00004.safetensors +3 -0
  6. model-00004-of-00004.safetensors +3 -0
  7. model.safetensors.index.json +754 -0
  8. trainer_state.json +767 -0
  9. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SvMoECLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 7168,
14
+ "max_position_embeddings": 4096,
15
+ "mlp_bias": false,
16
+ "model_type": "mixtral",
17
+ "num_attention_heads": 16,
18
+ "num_experts_per_tok": 2,
19
+ "num_hidden_layers": 24,
20
+ "num_key_value_heads": 16,
21
+ "num_local_experts": 8,
22
+ "output_router_logits": true,
23
+ "pretraining_tp": 1,
24
+ "rms_norm_eps": 1e-05,
25
+ "rope_scaling": null,
26
+ "rope_theta": 10000,
27
+ "router_aux_loss_coef": 0.01,
28
+ "router_jitter_noise": 0.0,
29
+ "sliding_window": null,
30
+ "tie_word_embeddings": false,
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.46.0",
33
+ "use_cache": true,
34
+ "vocab_size": 99584
35
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.46.0"
6
+ }
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:941bc6149d415d2c636c799aa1e38ae10926b26c25064bb0271ac472d25a1607
3
+ size 4988379392
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a72e41efe80fcdc58f61d7378d740eb49c05f0a4157e65636acfcb4c8eab7f7
3
+ size 4987299352
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d03db87f028feb6dcfe883b39d7b5b334753e3ac6534ca51142bc0d3efa5707c
3
+ size 4991535416
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:baa6bf25c224af17f9464daf0ea0c9aba532644b473d9ab4cc579351650a3453
3
+ size 3566400304
model.safetensors.index.json ADDED
@@ -0,0 +1,754 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 18533519360
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00004-of-00004.safetensors",
7
+ "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
8
+ "model.layers.0.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00004.safetensors",
9
+ "model.layers.0.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00004.safetensors",
10
+ "model.layers.0.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00004.safetensors",
11
+ "model.layers.0.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00004.safetensors",
12
+ "model.layers.0.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00004.safetensors",
13
+ "model.layers.0.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00004.safetensors",
14
+ "model.layers.0.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00004.safetensors",
15
+ "model.layers.0.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00004.safetensors",
16
+ "model.layers.0.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00004.safetensors",
17
+ "model.layers.0.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00004.safetensors",
18
+ "model.layers.0.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00004.safetensors",
19
+ "model.layers.0.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00004.safetensors",
20
+ "model.layers.0.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00004.safetensors",
21
+ "model.layers.0.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00004.safetensors",
22
+ "model.layers.0.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00004.safetensors",
23
+ "model.layers.0.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00004.safetensors",
24
+ "model.layers.0.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00004.safetensors",
25
+ "model.layers.0.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00004.safetensors",
26
+ "model.layers.0.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00004.safetensors",
27
+ "model.layers.0.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00004.safetensors",
28
+ "model.layers.0.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00004.safetensors",
29
+ "model.layers.0.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00004.safetensors",
30
+ "model.layers.0.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00004.safetensors",
31
+ "model.layers.0.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00004.safetensors",
32
+ "model.layers.0.block_sparse_moe.gate.weight": "model-00001-of-00004.safetensors",
33
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
34
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
35
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
36
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
37
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
38
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
39
+ "model.layers.1.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00004.safetensors",
40
+ "model.layers.1.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00004.safetensors",
41
+ "model.layers.1.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00004.safetensors",
42
+ "model.layers.1.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00004.safetensors",
43
+ "model.layers.1.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00004.safetensors",
44
+ "model.layers.1.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00004.safetensors",
45
+ "model.layers.1.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00004.safetensors",
46
+ "model.layers.1.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00004.safetensors",
47
+ "model.layers.1.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00004.safetensors",
48
+ "model.layers.1.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00004.safetensors",
49
+ "model.layers.1.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00004.safetensors",
50
+ "model.layers.1.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00004.safetensors",
51
+ "model.layers.1.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00004.safetensors",
52
+ "model.layers.1.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00004.safetensors",
53
+ "model.layers.1.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00004.safetensors",
54
+ "model.layers.1.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00004.safetensors",
55
+ "model.layers.1.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00004.safetensors",
56
+ "model.layers.1.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00004.safetensors",
57
+ "model.layers.1.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00004.safetensors",
58
+ "model.layers.1.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00004.safetensors",
59
+ "model.layers.1.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00004.safetensors",
60
+ "model.layers.1.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00004.safetensors",
61
+ "model.layers.1.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00004.safetensors",
62
+ "model.layers.1.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00004.safetensors",
63
+ "model.layers.1.block_sparse_moe.gate.weight": "model-00001-of-00004.safetensors",
64
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
65
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
66
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
67
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
68
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
69
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
70
+ "model.layers.10.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00004.safetensors",
71
+ "model.layers.10.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00004.safetensors",
72
+ "model.layers.10.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00004.safetensors",
73
+ "model.layers.10.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00004.safetensors",
74
+ "model.layers.10.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00004.safetensors",
75
+ "model.layers.10.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00004.safetensors",
76
+ "model.layers.10.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00004.safetensors",
77
+ "model.layers.10.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00004.safetensors",
78
+ "model.layers.10.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00004.safetensors",
79
+ "model.layers.10.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00004.safetensors",
80
+ "model.layers.10.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00004.safetensors",
81
+ "model.layers.10.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00004.safetensors",
82
+ "model.layers.10.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00004.safetensors",
83
+ "model.layers.10.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00004.safetensors",
84
+ "model.layers.10.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00004.safetensors",
85
+ "model.layers.10.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00004.safetensors",
86
+ "model.layers.10.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00004.safetensors",
87
+ "model.layers.10.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00004.safetensors",
88
+ "model.layers.10.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00004.safetensors",
89
+ "model.layers.10.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00004.safetensors",
90
+ "model.layers.10.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00004.safetensors",
91
+ "model.layers.10.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00004.safetensors",
92
+ "model.layers.10.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00004.safetensors",
93
+ "model.layers.10.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00004.safetensors",
94
+ "model.layers.10.block_sparse_moe.gate.weight": "model-00002-of-00004.safetensors",
95
+ "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
96
+ "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
97
+ "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
98
+ "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
99
+ "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
100
+ "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
101
+ "model.layers.11.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00004.safetensors",
102
+ "model.layers.11.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00004.safetensors",
103
+ "model.layers.11.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00004.safetensors",
104
+ "model.layers.11.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00004.safetensors",
105
+ "model.layers.11.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00004.safetensors",
106
+ "model.layers.11.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00004.safetensors",
107
+ "model.layers.11.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00004.safetensors",
108
+ "model.layers.11.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00004.safetensors",
109
+ "model.layers.11.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00004.safetensors",
110
+ "model.layers.11.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00004.safetensors",
111
+ "model.layers.11.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00004.safetensors",
112
+ "model.layers.11.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00004.safetensors",
113
+ "model.layers.11.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00004.safetensors",
114
+ "model.layers.11.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00004.safetensors",
115
+ "model.layers.11.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00004.safetensors",
116
+ "model.layers.11.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00004.safetensors",
117
+ "model.layers.11.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00004.safetensors",
118
+ "model.layers.11.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00004.safetensors",
119
+ "model.layers.11.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00004.safetensors",
120
+ "model.layers.11.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00004.safetensors",
121
+ "model.layers.11.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00004.safetensors",
122
+ "model.layers.11.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00004.safetensors",
123
+ "model.layers.11.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00004.safetensors",
124
+ "model.layers.11.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00004.safetensors",
125
+ "model.layers.11.block_sparse_moe.gate.weight": "model-00002-of-00004.safetensors",
126
+ "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
127
+ "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
128
+ "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
129
+ "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
130
+ "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
131
+ "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
132
+ "model.layers.12.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00004.safetensors",
133
+ "model.layers.12.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00004.safetensors",
134
+ "model.layers.12.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00004.safetensors",
135
+ "model.layers.12.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00004.safetensors",
136
+ "model.layers.12.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00004.safetensors",
137
+ "model.layers.12.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00004.safetensors",
138
+ "model.layers.12.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00004.safetensors",
139
+ "model.layers.12.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00004.safetensors",
140
+ "model.layers.12.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00004.safetensors",
141
+ "model.layers.12.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00004.safetensors",
142
+ "model.layers.12.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00004.safetensors",
143
+ "model.layers.12.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00004.safetensors",
144
+ "model.layers.12.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00004.safetensors",
145
+ "model.layers.12.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00004.safetensors",
146
+ "model.layers.12.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00004.safetensors",
147
+ "model.layers.12.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00004.safetensors",
148
+ "model.layers.12.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00004.safetensors",
149
+ "model.layers.12.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00004.safetensors",
150
+ "model.layers.12.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00004.safetensors",
151
+ "model.layers.12.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00004.safetensors",
152
+ "model.layers.12.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00004.safetensors",
153
+ "model.layers.12.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00004.safetensors",
154
+ "model.layers.12.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00004.safetensors",
155
+ "model.layers.12.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00004.safetensors",
156
+ "model.layers.12.block_sparse_moe.gate.weight": "model-00002-of-00004.safetensors",
157
+ "model.layers.12.input_layernorm.weight": "model-00003-of-00004.safetensors",
158
+ "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
159
+ "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
160
+ "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
161
+ "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
162
+ "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
163
+ "model.layers.13.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00004.safetensors",
164
+ "model.layers.13.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00004.safetensors",
165
+ "model.layers.13.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00004.safetensors",
166
+ "model.layers.13.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00004.safetensors",
167
+ "model.layers.13.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00004.safetensors",
168
+ "model.layers.13.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00004.safetensors",
169
+ "model.layers.13.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00004.safetensors",
170
+ "model.layers.13.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00004.safetensors",
171
+ "model.layers.13.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00004.safetensors",
172
+ "model.layers.13.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00004.safetensors",
173
+ "model.layers.13.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00004.safetensors",
174
+ "model.layers.13.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00004.safetensors",
175
+ "model.layers.13.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00004.safetensors",
176
+ "model.layers.13.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00004.safetensors",
177
+ "model.layers.13.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00004.safetensors",
178
+ "model.layers.13.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00004.safetensors",
179
+ "model.layers.13.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00004.safetensors",
180
+ "model.layers.13.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00004.safetensors",
181
+ "model.layers.13.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00004.safetensors",
182
+ "model.layers.13.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00004.safetensors",
183
+ "model.layers.13.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00004.safetensors",
184
+ "model.layers.13.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00004.safetensors",
185
+ "model.layers.13.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00004.safetensors",
186
+ "model.layers.13.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00004.safetensors",
187
+ "model.layers.13.block_sparse_moe.gate.weight": "model-00003-of-00004.safetensors",
188
+ "model.layers.13.input_layernorm.weight": "model-00003-of-00004.safetensors",
189
+ "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
190
+ "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
191
+ "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
192
+ "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
193
+ "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
194
+ "model.layers.14.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00004.safetensors",
195
+ "model.layers.14.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00004.safetensors",
196
+ "model.layers.14.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00004.safetensors",
197
+ "model.layers.14.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00004.safetensors",
198
+ "model.layers.14.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00004.safetensors",
199
+ "model.layers.14.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00004.safetensors",
200
+ "model.layers.14.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00004.safetensors",
201
+ "model.layers.14.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00004.safetensors",
202
+ "model.layers.14.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00004.safetensors",
203
+ "model.layers.14.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00004.safetensors",
204
+ "model.layers.14.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00004.safetensors",
205
+ "model.layers.14.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00004.safetensors",
206
+ "model.layers.14.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00004.safetensors",
207
+ "model.layers.14.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00004.safetensors",
208
+ "model.layers.14.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00004.safetensors",
209
+ "model.layers.14.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00004.safetensors",
210
+ "model.layers.14.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00004.safetensors",
211
+ "model.layers.14.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00004.safetensors",
212
+ "model.layers.14.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00004.safetensors",
213
+ "model.layers.14.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00004.safetensors",
214
+ "model.layers.14.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00004.safetensors",
215
+ "model.layers.14.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00004.safetensors",
216
+ "model.layers.14.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00004.safetensors",
217
+ "model.layers.14.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00004.safetensors",
218
+ "model.layers.14.block_sparse_moe.gate.weight": "model-00003-of-00004.safetensors",
219
+ "model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
220
+ "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
221
+ "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
222
+ "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
223
+ "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
224
+ "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
225
+ "model.layers.15.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00004.safetensors",
226
+ "model.layers.15.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00004.safetensors",
227
+ "model.layers.15.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00004.safetensors",
228
+ "model.layers.15.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00004.safetensors",
229
+ "model.layers.15.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00004.safetensors",
230
+ "model.layers.15.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00004.safetensors",
231
+ "model.layers.15.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00004.safetensors",
232
+ "model.layers.15.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00004.safetensors",
233
+ "model.layers.15.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00004.safetensors",
234
+ "model.layers.15.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00004.safetensors",
235
+ "model.layers.15.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00004.safetensors",
236
+ "model.layers.15.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00004.safetensors",
237
+ "model.layers.15.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00004.safetensors",
238
+ "model.layers.15.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00004.safetensors",
239
+ "model.layers.15.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00004.safetensors",
240
+ "model.layers.15.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00004.safetensors",
241
+ "model.layers.15.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00004.safetensors",
242
+ "model.layers.15.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00004.safetensors",
243
+ "model.layers.15.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00004.safetensors",
244
+ "model.layers.15.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00004.safetensors",
245
+ "model.layers.15.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00004.safetensors",
246
+ "model.layers.15.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00004.safetensors",
247
+ "model.layers.15.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00004.safetensors",
248
+ "model.layers.15.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00004.safetensors",
249
+ "model.layers.15.block_sparse_moe.gate.weight": "model-00003-of-00004.safetensors",
250
+ "model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
251
+ "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
252
+ "model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
253
+ "model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
254
+ "model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
255
+ "model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
256
+ "model.layers.16.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00004.safetensors",
257
+ "model.layers.16.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00004.safetensors",
258
+ "model.layers.16.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00004.safetensors",
259
+ "model.layers.16.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00004.safetensors",
260
+ "model.layers.16.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00004.safetensors",
261
+ "model.layers.16.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00004.safetensors",
262
+ "model.layers.16.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00004.safetensors",
263
+ "model.layers.16.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00004.safetensors",
264
+ "model.layers.16.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00004.safetensors",
265
+ "model.layers.16.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00004.safetensors",
266
+ "model.layers.16.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00004.safetensors",
267
+ "model.layers.16.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00004.safetensors",
268
+ "model.layers.16.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00004.safetensors",
269
+ "model.layers.16.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00004.safetensors",
270
+ "model.layers.16.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00004.safetensors",
271
+ "model.layers.16.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00004.safetensors",
272
+ "model.layers.16.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00004.safetensors",
273
+ "model.layers.16.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00004.safetensors",
274
+ "model.layers.16.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00004.safetensors",
275
+ "model.layers.16.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00004.safetensors",
276
+ "model.layers.16.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00004.safetensors",
277
+ "model.layers.16.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00004.safetensors",
278
+ "model.layers.16.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00004.safetensors",
279
+ "model.layers.16.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00004.safetensors",
280
+ "model.layers.16.block_sparse_moe.gate.weight": "model-00003-of-00004.safetensors",
281
+ "model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
282
+ "model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
283
+ "model.layers.16.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
284
+ "model.layers.16.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
285
+ "model.layers.16.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
286
+ "model.layers.16.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
287
+ "model.layers.17.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00004.safetensors",
288
+ "model.layers.17.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00004.safetensors",
289
+ "model.layers.17.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00004.safetensors",
290
+ "model.layers.17.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00004.safetensors",
291
+ "model.layers.17.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00004.safetensors",
292
+ "model.layers.17.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00004.safetensors",
293
+ "model.layers.17.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00004.safetensors",
294
+ "model.layers.17.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00004.safetensors",
295
+ "model.layers.17.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00004.safetensors",
296
+ "model.layers.17.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00004.safetensors",
297
+ "model.layers.17.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00004.safetensors",
298
+ "model.layers.17.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00004.safetensors",
299
+ "model.layers.17.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00004.safetensors",
300
+ "model.layers.17.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00004.safetensors",
301
+ "model.layers.17.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00004.safetensors",
302
+ "model.layers.17.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00004.safetensors",
303
+ "model.layers.17.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00004.safetensors",
304
+ "model.layers.17.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00004.safetensors",
305
+ "model.layers.17.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00004.safetensors",
306
+ "model.layers.17.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00004.safetensors",
307
+ "model.layers.17.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00004.safetensors",
308
+ "model.layers.17.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00004.safetensors",
309
+ "model.layers.17.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00004.safetensors",
310
+ "model.layers.17.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00004.safetensors",
311
+ "model.layers.17.block_sparse_moe.gate.weight": "model-00003-of-00004.safetensors",
312
+ "model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
313
+ "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
314
+ "model.layers.17.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
315
+ "model.layers.17.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
316
+ "model.layers.17.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
317
+ "model.layers.17.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
318
+ "model.layers.18.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00004.safetensors",
319
+ "model.layers.18.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00004.safetensors",
320
+ "model.layers.18.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00004.safetensors",
321
+ "model.layers.18.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00004.safetensors",
322
+ "model.layers.18.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00004.safetensors",
323
+ "model.layers.18.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00004.safetensors",
324
+ "model.layers.18.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00004.safetensors",
325
+ "model.layers.18.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00004.safetensors",
326
+ "model.layers.18.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00004.safetensors",
327
+ "model.layers.18.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00004.safetensors",
328
+ "model.layers.18.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00004.safetensors",
329
+ "model.layers.18.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00004.safetensors",
330
+ "model.layers.18.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00004.safetensors",
331
+ "model.layers.18.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00004.safetensors",
332
+ "model.layers.18.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00004.safetensors",
333
+ "model.layers.18.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00004.safetensors",
334
+ "model.layers.18.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00004.safetensors",
335
+ "model.layers.18.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00004.safetensors",
336
+ "model.layers.18.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00004.safetensors",
337
+ "model.layers.18.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00004.safetensors",
338
+ "model.layers.18.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00004.safetensors",
339
+ "model.layers.18.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00004.safetensors",
340
+ "model.layers.18.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00004.safetensors",
341
+ "model.layers.18.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00004.safetensors",
342
+ "model.layers.18.block_sparse_moe.gate.weight": "model-00003-of-00004.safetensors",
343
+ "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
344
+ "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
345
+ "model.layers.18.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
346
+ "model.layers.18.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
347
+ "model.layers.18.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
348
+ "model.layers.18.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
349
+ "model.layers.19.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00004.safetensors",
350
+ "model.layers.19.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00004.safetensors",
351
+ "model.layers.19.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00004.safetensors",
352
+ "model.layers.19.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00004.safetensors",
353
+ "model.layers.19.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00004.safetensors",
354
+ "model.layers.19.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00004.safetensors",
355
+ "model.layers.19.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00004.safetensors",
356
+ "model.layers.19.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00004.safetensors",
357
+ "model.layers.19.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00004.safetensors",
358
+ "model.layers.19.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00004.safetensors",
359
+ "model.layers.19.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00004.safetensors",
360
+ "model.layers.19.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00004.safetensors",
361
+ "model.layers.19.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00004.safetensors",
362
+ "model.layers.19.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00004.safetensors",
363
+ "model.layers.19.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00004.safetensors",
364
+ "model.layers.19.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00004.safetensors",
365
+ "model.layers.19.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00004.safetensors",
366
+ "model.layers.19.block_sparse_moe.experts.5.w3.weight": "model-00004-of-00004.safetensors",
367
+ "model.layers.19.block_sparse_moe.experts.6.w1.weight": "model-00004-of-00004.safetensors",
368
+ "model.layers.19.block_sparse_moe.experts.6.w2.weight": "model-00004-of-00004.safetensors",
369
+ "model.layers.19.block_sparse_moe.experts.6.w3.weight": "model-00004-of-00004.safetensors",
370
+ "model.layers.19.block_sparse_moe.experts.7.w1.weight": "model-00004-of-00004.safetensors",
371
+ "model.layers.19.block_sparse_moe.experts.7.w2.weight": "model-00004-of-00004.safetensors",
372
+ "model.layers.19.block_sparse_moe.experts.7.w3.weight": "model-00004-of-00004.safetensors",
373
+ "model.layers.19.block_sparse_moe.gate.weight": "model-00003-of-00004.safetensors",
374
+ "model.layers.19.input_layernorm.weight": "model-00004-of-00004.safetensors",
375
+ "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
376
+ "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
377
+ "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
378
+ "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
379
+ "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
380
+ "model.layers.2.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00004.safetensors",
381
+ "model.layers.2.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00004.safetensors",
382
+ "model.layers.2.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00004.safetensors",
383
+ "model.layers.2.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00004.safetensors",
384
+ "model.layers.2.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00004.safetensors",
385
+ "model.layers.2.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00004.safetensors",
386
+ "model.layers.2.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00004.safetensors",
387
+ "model.layers.2.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00004.safetensors",
388
+ "model.layers.2.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00004.safetensors",
389
+ "model.layers.2.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00004.safetensors",
390
+ "model.layers.2.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00004.safetensors",
391
+ "model.layers.2.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00004.safetensors",
392
+ "model.layers.2.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00004.safetensors",
393
+ "model.layers.2.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00004.safetensors",
394
+ "model.layers.2.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00004.safetensors",
395
+ "model.layers.2.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00004.safetensors",
396
+ "model.layers.2.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00004.safetensors",
397
+ "model.layers.2.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00004.safetensors",
398
+ "model.layers.2.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00004.safetensors",
399
+ "model.layers.2.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00004.safetensors",
400
+ "model.layers.2.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00004.safetensors",
401
+ "model.layers.2.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00004.safetensors",
402
+ "model.layers.2.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00004.safetensors",
403
+ "model.layers.2.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00004.safetensors",
404
+ "model.layers.2.block_sparse_moe.gate.weight": "model-00001-of-00004.safetensors",
405
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
406
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
407
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
408
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
409
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
410
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
411
+ "model.layers.20.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00004.safetensors",
412
+ "model.layers.20.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00004.safetensors",
413
+ "model.layers.20.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00004.safetensors",
414
+ "model.layers.20.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00004.safetensors",
415
+ "model.layers.20.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00004.safetensors",
416
+ "model.layers.20.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00004.safetensors",
417
+ "model.layers.20.block_sparse_moe.experts.2.w1.weight": "model-00004-of-00004.safetensors",
418
+ "model.layers.20.block_sparse_moe.experts.2.w2.weight": "model-00004-of-00004.safetensors",
419
+ "model.layers.20.block_sparse_moe.experts.2.w3.weight": "model-00004-of-00004.safetensors",
420
+ "model.layers.20.block_sparse_moe.experts.3.w1.weight": "model-00004-of-00004.safetensors",
421
+ "model.layers.20.block_sparse_moe.experts.3.w2.weight": "model-00004-of-00004.safetensors",
422
+ "model.layers.20.block_sparse_moe.experts.3.w3.weight": "model-00004-of-00004.safetensors",
423
+ "model.layers.20.block_sparse_moe.experts.4.w1.weight": "model-00004-of-00004.safetensors",
424
+ "model.layers.20.block_sparse_moe.experts.4.w2.weight": "model-00004-of-00004.safetensors",
425
+ "model.layers.20.block_sparse_moe.experts.4.w3.weight": "model-00004-of-00004.safetensors",
426
+ "model.layers.20.block_sparse_moe.experts.5.w1.weight": "model-00004-of-00004.safetensors",
427
+ "model.layers.20.block_sparse_moe.experts.5.w2.weight": "model-00004-of-00004.safetensors",
428
+ "model.layers.20.block_sparse_moe.experts.5.w3.weight": "model-00004-of-00004.safetensors",
429
+ "model.layers.20.block_sparse_moe.experts.6.w1.weight": "model-00004-of-00004.safetensors",
430
+ "model.layers.20.block_sparse_moe.experts.6.w2.weight": "model-00004-of-00004.safetensors",
431
+ "model.layers.20.block_sparse_moe.experts.6.w3.weight": "model-00004-of-00004.safetensors",
432
+ "model.layers.20.block_sparse_moe.experts.7.w1.weight": "model-00004-of-00004.safetensors",
433
+ "model.layers.20.block_sparse_moe.experts.7.w2.weight": "model-00004-of-00004.safetensors",
434
+ "model.layers.20.block_sparse_moe.experts.7.w3.weight": "model-00004-of-00004.safetensors",
435
+ "model.layers.20.block_sparse_moe.gate.weight": "model-00004-of-00004.safetensors",
436
+ "model.layers.20.input_layernorm.weight": "model-00004-of-00004.safetensors",
437
+ "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
438
+ "model.layers.20.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
439
+ "model.layers.20.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
440
+ "model.layers.20.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
441
+ "model.layers.20.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
442
+ "model.layers.21.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00004.safetensors",
443
+ "model.layers.21.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00004.safetensors",
444
+ "model.layers.21.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00004.safetensors",
445
+ "model.layers.21.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00004.safetensors",
446
+ "model.layers.21.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00004.safetensors",
447
+ "model.layers.21.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00004.safetensors",
448
+ "model.layers.21.block_sparse_moe.experts.2.w1.weight": "model-00004-of-00004.safetensors",
449
+ "model.layers.21.block_sparse_moe.experts.2.w2.weight": "model-00004-of-00004.safetensors",
450
+ "model.layers.21.block_sparse_moe.experts.2.w3.weight": "model-00004-of-00004.safetensors",
451
+ "model.layers.21.block_sparse_moe.experts.3.w1.weight": "model-00004-of-00004.safetensors",
452
+ "model.layers.21.block_sparse_moe.experts.3.w2.weight": "model-00004-of-00004.safetensors",
453
+ "model.layers.21.block_sparse_moe.experts.3.w3.weight": "model-00004-of-00004.safetensors",
454
+ "model.layers.21.block_sparse_moe.experts.4.w1.weight": "model-00004-of-00004.safetensors",
455
+ "model.layers.21.block_sparse_moe.experts.4.w2.weight": "model-00004-of-00004.safetensors",
456
+ "model.layers.21.block_sparse_moe.experts.4.w3.weight": "model-00004-of-00004.safetensors",
457
+ "model.layers.21.block_sparse_moe.experts.5.w1.weight": "model-00004-of-00004.safetensors",
458
+ "model.layers.21.block_sparse_moe.experts.5.w2.weight": "model-00004-of-00004.safetensors",
459
+ "model.layers.21.block_sparse_moe.experts.5.w3.weight": "model-00004-of-00004.safetensors",
460
+ "model.layers.21.block_sparse_moe.experts.6.w1.weight": "model-00004-of-00004.safetensors",
461
+ "model.layers.21.block_sparse_moe.experts.6.w2.weight": "model-00004-of-00004.safetensors",
462
+ "model.layers.21.block_sparse_moe.experts.6.w3.weight": "model-00004-of-00004.safetensors",
463
+ "model.layers.21.block_sparse_moe.experts.7.w1.weight": "model-00004-of-00004.safetensors",
464
+ "model.layers.21.block_sparse_moe.experts.7.w2.weight": "model-00004-of-00004.safetensors",
465
+ "model.layers.21.block_sparse_moe.experts.7.w3.weight": "model-00004-of-00004.safetensors",
466
+ "model.layers.21.block_sparse_moe.gate.weight": "model-00004-of-00004.safetensors",
467
+ "model.layers.21.input_layernorm.weight": "model-00004-of-00004.safetensors",
468
+ "model.layers.21.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
469
+ "model.layers.21.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
470
+ "model.layers.21.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
471
+ "model.layers.21.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
472
+ "model.layers.21.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
473
+ "model.layers.22.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00004.safetensors",
474
+ "model.layers.22.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00004.safetensors",
475
+ "model.layers.22.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00004.safetensors",
476
+ "model.layers.22.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00004.safetensors",
477
+ "model.layers.22.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00004.safetensors",
478
+ "model.layers.22.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00004.safetensors",
479
+ "model.layers.22.block_sparse_moe.experts.2.w1.weight": "model-00004-of-00004.safetensors",
480
+ "model.layers.22.block_sparse_moe.experts.2.w2.weight": "model-00004-of-00004.safetensors",
481
+ "model.layers.22.block_sparse_moe.experts.2.w3.weight": "model-00004-of-00004.safetensors",
482
+ "model.layers.22.block_sparse_moe.experts.3.w1.weight": "model-00004-of-00004.safetensors",
483
+ "model.layers.22.block_sparse_moe.experts.3.w2.weight": "model-00004-of-00004.safetensors",
484
+ "model.layers.22.block_sparse_moe.experts.3.w3.weight": "model-00004-of-00004.safetensors",
485
+ "model.layers.22.block_sparse_moe.experts.4.w1.weight": "model-00004-of-00004.safetensors",
486
+ "model.layers.22.block_sparse_moe.experts.4.w2.weight": "model-00004-of-00004.safetensors",
487
+ "model.layers.22.block_sparse_moe.experts.4.w3.weight": "model-00004-of-00004.safetensors",
488
+ "model.layers.22.block_sparse_moe.experts.5.w1.weight": "model-00004-of-00004.safetensors",
489
+ "model.layers.22.block_sparse_moe.experts.5.w2.weight": "model-00004-of-00004.safetensors",
490
+ "model.layers.22.block_sparse_moe.experts.5.w3.weight": "model-00004-of-00004.safetensors",
491
+ "model.layers.22.block_sparse_moe.experts.6.w1.weight": "model-00004-of-00004.safetensors",
492
+ "model.layers.22.block_sparse_moe.experts.6.w2.weight": "model-00004-of-00004.safetensors",
493
+ "model.layers.22.block_sparse_moe.experts.6.w3.weight": "model-00004-of-00004.safetensors",
494
+ "model.layers.22.block_sparse_moe.experts.7.w1.weight": "model-00004-of-00004.safetensors",
495
+ "model.layers.22.block_sparse_moe.experts.7.w2.weight": "model-00004-of-00004.safetensors",
496
+ "model.layers.22.block_sparse_moe.experts.7.w3.weight": "model-00004-of-00004.safetensors",
497
+ "model.layers.22.block_sparse_moe.gate.weight": "model-00004-of-00004.safetensors",
498
+ "model.layers.22.input_layernorm.weight": "model-00004-of-00004.safetensors",
499
+ "model.layers.22.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
500
+ "model.layers.22.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
501
+ "model.layers.22.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
502
+ "model.layers.22.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
503
+ "model.layers.22.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
504
+ "model.layers.23.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00004.safetensors",
505
+ "model.layers.23.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00004.safetensors",
506
+ "model.layers.23.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00004.safetensors",
507
+ "model.layers.23.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00004.safetensors",
508
+ "model.layers.23.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00004.safetensors",
509
+ "model.layers.23.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00004.safetensors",
510
+ "model.layers.23.block_sparse_moe.experts.2.w1.weight": "model-00004-of-00004.safetensors",
511
+ "model.layers.23.block_sparse_moe.experts.2.w2.weight": "model-00004-of-00004.safetensors",
512
+ "model.layers.23.block_sparse_moe.experts.2.w3.weight": "model-00004-of-00004.safetensors",
513
+ "model.layers.23.block_sparse_moe.experts.3.w1.weight": "model-00004-of-00004.safetensors",
514
+ "model.layers.23.block_sparse_moe.experts.3.w2.weight": "model-00004-of-00004.safetensors",
515
+ "model.layers.23.block_sparse_moe.experts.3.w3.weight": "model-00004-of-00004.safetensors",
516
+ "model.layers.23.block_sparse_moe.experts.4.w1.weight": "model-00004-of-00004.safetensors",
517
+ "model.layers.23.block_sparse_moe.experts.4.w2.weight": "model-00004-of-00004.safetensors",
518
+ "model.layers.23.block_sparse_moe.experts.4.w3.weight": "model-00004-of-00004.safetensors",
519
+ "model.layers.23.block_sparse_moe.experts.5.w1.weight": "model-00004-of-00004.safetensors",
520
+ "model.layers.23.block_sparse_moe.experts.5.w2.weight": "model-00004-of-00004.safetensors",
521
+ "model.layers.23.block_sparse_moe.experts.5.w3.weight": "model-00004-of-00004.safetensors",
522
+ "model.layers.23.block_sparse_moe.experts.6.w1.weight": "model-00004-of-00004.safetensors",
523
+ "model.layers.23.block_sparse_moe.experts.6.w2.weight": "model-00004-of-00004.safetensors",
524
+ "model.layers.23.block_sparse_moe.experts.6.w3.weight": "model-00004-of-00004.safetensors",
525
+ "model.layers.23.block_sparse_moe.experts.7.w1.weight": "model-00004-of-00004.safetensors",
526
+ "model.layers.23.block_sparse_moe.experts.7.w2.weight": "model-00004-of-00004.safetensors",
527
+ "model.layers.23.block_sparse_moe.experts.7.w3.weight": "model-00004-of-00004.safetensors",
528
+ "model.layers.23.block_sparse_moe.gate.weight": "model-00004-of-00004.safetensors",
529
+ "model.layers.23.input_layernorm.weight": "model-00004-of-00004.safetensors",
530
+ "model.layers.23.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
531
+ "model.layers.23.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
532
+ "model.layers.23.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
533
+ "model.layers.23.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
534
+ "model.layers.23.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
535
+ "model.layers.3.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00004.safetensors",
536
+ "model.layers.3.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00004.safetensors",
537
+ "model.layers.3.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00004.safetensors",
538
+ "model.layers.3.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00004.safetensors",
539
+ "model.layers.3.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00004.safetensors",
540
+ "model.layers.3.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00004.safetensors",
541
+ "model.layers.3.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00004.safetensors",
542
+ "model.layers.3.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00004.safetensors",
543
+ "model.layers.3.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00004.safetensors",
544
+ "model.layers.3.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00004.safetensors",
545
+ "model.layers.3.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00004.safetensors",
546
+ "model.layers.3.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00004.safetensors",
547
+ "model.layers.3.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00004.safetensors",
548
+ "model.layers.3.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00004.safetensors",
549
+ "model.layers.3.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00004.safetensors",
550
+ "model.layers.3.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00004.safetensors",
551
+ "model.layers.3.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00004.safetensors",
552
+ "model.layers.3.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00004.safetensors",
553
+ "model.layers.3.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00004.safetensors",
554
+ "model.layers.3.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00004.safetensors",
555
+ "model.layers.3.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00004.safetensors",
556
+ "model.layers.3.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00004.safetensors",
557
+ "model.layers.3.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00004.safetensors",
558
+ "model.layers.3.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00004.safetensors",
559
+ "model.layers.3.block_sparse_moe.gate.weight": "model-00001-of-00004.safetensors",
560
+ "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
561
+ "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
562
+ "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
563
+ "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
564
+ "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
565
+ "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
566
+ "model.layers.4.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00004.safetensors",
567
+ "model.layers.4.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00004.safetensors",
568
+ "model.layers.4.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00004.safetensors",
569
+ "model.layers.4.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00004.safetensors",
570
+ "model.layers.4.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00004.safetensors",
571
+ "model.layers.4.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00004.safetensors",
572
+ "model.layers.4.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00004.safetensors",
573
+ "model.layers.4.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00004.safetensors",
574
+ "model.layers.4.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00004.safetensors",
575
+ "model.layers.4.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00004.safetensors",
576
+ "model.layers.4.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00004.safetensors",
577
+ "model.layers.4.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00004.safetensors",
578
+ "model.layers.4.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00004.safetensors",
579
+ "model.layers.4.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00004.safetensors",
580
+ "model.layers.4.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00004.safetensors",
581
+ "model.layers.4.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00004.safetensors",
582
+ "model.layers.4.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00004.safetensors",
583
+ "model.layers.4.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00004.safetensors",
584
+ "model.layers.4.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00004.safetensors",
585
+ "model.layers.4.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00004.safetensors",
586
+ "model.layers.4.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00004.safetensors",
587
+ "model.layers.4.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00004.safetensors",
588
+ "model.layers.4.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00004.safetensors",
589
+ "model.layers.4.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00004.safetensors",
590
+ "model.layers.4.block_sparse_moe.gate.weight": "model-00001-of-00004.safetensors",
591
+ "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
592
+ "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
593
+ "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
594
+ "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
595
+ "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
596
+ "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
597
+ "model.layers.5.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00004.safetensors",
598
+ "model.layers.5.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00004.safetensors",
599
+ "model.layers.5.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00004.safetensors",
600
+ "model.layers.5.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00004.safetensors",
601
+ "model.layers.5.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00004.safetensors",
602
+ "model.layers.5.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00004.safetensors",
603
+ "model.layers.5.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00004.safetensors",
604
+ "model.layers.5.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00004.safetensors",
605
+ "model.layers.5.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00004.safetensors",
606
+ "model.layers.5.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00004.safetensors",
607
+ "model.layers.5.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00004.safetensors",
608
+ "model.layers.5.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00004.safetensors",
609
+ "model.layers.5.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00004.safetensors",
610
+ "model.layers.5.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00004.safetensors",
611
+ "model.layers.5.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00004.safetensors",
612
+ "model.layers.5.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00004.safetensors",
613
+ "model.layers.5.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00004.safetensors",
614
+ "model.layers.5.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00004.safetensors",
615
+ "model.layers.5.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00004.safetensors",
616
+ "model.layers.5.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00004.safetensors",
617
+ "model.layers.5.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00004.safetensors",
618
+ "model.layers.5.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00004.safetensors",
619
+ "model.layers.5.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00004.safetensors",
620
+ "model.layers.5.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00004.safetensors",
621
+ "model.layers.5.block_sparse_moe.gate.weight": "model-00001-of-00004.safetensors",
622
+ "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
623
+ "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
624
+ "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
625
+ "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
626
+ "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
627
+ "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
628
+ "model.layers.6.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00004.safetensors",
629
+ "model.layers.6.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00004.safetensors",
630
+ "model.layers.6.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00004.safetensors",
631
+ "model.layers.6.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00004.safetensors",
632
+ "model.layers.6.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00004.safetensors",
633
+ "model.layers.6.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00004.safetensors",
634
+ "model.layers.6.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00004.safetensors",
635
+ "model.layers.6.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00004.safetensors",
636
+ "model.layers.6.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00004.safetensors",
637
+ "model.layers.6.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00004.safetensors",
638
+ "model.layers.6.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00004.safetensors",
639
+ "model.layers.6.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00004.safetensors",
640
+ "model.layers.6.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00004.safetensors",
641
+ "model.layers.6.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00004.safetensors",
642
+ "model.layers.6.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00004.safetensors",
643
+ "model.layers.6.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00004.safetensors",
644
+ "model.layers.6.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00004.safetensors",
645
+ "model.layers.6.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00004.safetensors",
646
+ "model.layers.6.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00004.safetensors",
647
+ "model.layers.6.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00004.safetensors",
648
+ "model.layers.6.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00004.safetensors",
649
+ "model.layers.6.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00004.safetensors",
650
+ "model.layers.6.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00004.safetensors",
651
+ "model.layers.6.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00004.safetensors",
652
+ "model.layers.6.block_sparse_moe.gate.weight": "model-00001-of-00004.safetensors",
653
+ "model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
654
+ "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
655
+ "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
656
+ "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
657
+ "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
658
+ "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
659
+ "model.layers.7.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00004.safetensors",
660
+ "model.layers.7.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00004.safetensors",
661
+ "model.layers.7.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00004.safetensors",
662
+ "model.layers.7.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00004.safetensors",
663
+ "model.layers.7.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00004.safetensors",
664
+ "model.layers.7.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00004.safetensors",
665
+ "model.layers.7.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00004.safetensors",
666
+ "model.layers.7.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00004.safetensors",
667
+ "model.layers.7.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00004.safetensors",
668
+ "model.layers.7.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00004.safetensors",
669
+ "model.layers.7.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00004.safetensors",
670
+ "model.layers.7.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00004.safetensors",
671
+ "model.layers.7.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00004.safetensors",
672
+ "model.layers.7.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00004.safetensors",
673
+ "model.layers.7.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00004.safetensors",
674
+ "model.layers.7.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00004.safetensors",
675
+ "model.layers.7.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00004.safetensors",
676
+ "model.layers.7.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00004.safetensors",
677
+ "model.layers.7.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00004.safetensors",
678
+ "model.layers.7.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00004.safetensors",
679
+ "model.layers.7.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00004.safetensors",
680
+ "model.layers.7.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00004.safetensors",
681
+ "model.layers.7.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00004.safetensors",
682
+ "model.layers.7.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00004.safetensors",
683
+ "model.layers.7.block_sparse_moe.gate.weight": "model-00002-of-00004.safetensors",
684
+ "model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
685
+ "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
686
+ "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
687
+ "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
688
+ "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
689
+ "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
690
+ "model.layers.8.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00004.safetensors",
691
+ "model.layers.8.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00004.safetensors",
692
+ "model.layers.8.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00004.safetensors",
693
+ "model.layers.8.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00004.safetensors",
694
+ "model.layers.8.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00004.safetensors",
695
+ "model.layers.8.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00004.safetensors",
696
+ "model.layers.8.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00004.safetensors",
697
+ "model.layers.8.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00004.safetensors",
698
+ "model.layers.8.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00004.safetensors",
699
+ "model.layers.8.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00004.safetensors",
700
+ "model.layers.8.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00004.safetensors",
701
+ "model.layers.8.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00004.safetensors",
702
+ "model.layers.8.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00004.safetensors",
703
+ "model.layers.8.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00004.safetensors",
704
+ "model.layers.8.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00004.safetensors",
705
+ "model.layers.8.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00004.safetensors",
706
+ "model.layers.8.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00004.safetensors",
707
+ "model.layers.8.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00004.safetensors",
708
+ "model.layers.8.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00004.safetensors",
709
+ "model.layers.8.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00004.safetensors",
710
+ "model.layers.8.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00004.safetensors",
711
+ "model.layers.8.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00004.safetensors",
712
+ "model.layers.8.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00004.safetensors",
713
+ "model.layers.8.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00004.safetensors",
714
+ "model.layers.8.block_sparse_moe.gate.weight": "model-00002-of-00004.safetensors",
715
+ "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
716
+ "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
717
+ "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
718
+ "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
719
+ "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
720
+ "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
721
+ "model.layers.9.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00004.safetensors",
722
+ "model.layers.9.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00004.safetensors",
723
+ "model.layers.9.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00004.safetensors",
724
+ "model.layers.9.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00004.safetensors",
725
+ "model.layers.9.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00004.safetensors",
726
+ "model.layers.9.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00004.safetensors",
727
+ "model.layers.9.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00004.safetensors",
728
+ "model.layers.9.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00004.safetensors",
729
+ "model.layers.9.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00004.safetensors",
730
+ "model.layers.9.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00004.safetensors",
731
+ "model.layers.9.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00004.safetensors",
732
+ "model.layers.9.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00004.safetensors",
733
+ "model.layers.9.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00004.safetensors",
734
+ "model.layers.9.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00004.safetensors",
735
+ "model.layers.9.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00004.safetensors",
736
+ "model.layers.9.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00004.safetensors",
737
+ "model.layers.9.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00004.safetensors",
738
+ "model.layers.9.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00004.safetensors",
739
+ "model.layers.9.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00004.safetensors",
740
+ "model.layers.9.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00004.safetensors",
741
+ "model.layers.9.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00004.safetensors",
742
+ "model.layers.9.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00004.safetensors",
743
+ "model.layers.9.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00004.safetensors",
744
+ "model.layers.9.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00004.safetensors",
745
+ "model.layers.9.block_sparse_moe.gate.weight": "model-00002-of-00004.safetensors",
746
+ "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
747
+ "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
748
+ "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
749
+ "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
750
+ "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
751
+ "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
752
+ "model.norm.weight": "model-00004-of-00004.safetensors"
753
+ }
754
+ }
trainer_state.json ADDED
@@ -0,0 +1,767 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.21310460567474365,
3
+ "best_model_checkpoint": "../../models/baseline/btx/checkpoint-13692",
4
+ "epoch": 1.0,
5
+ "eval_steps": 3423,
6
+ "global_step": 13692,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.010005842827928718,
13
+ "grad_norm": 2.8703060150146484,
14
+ "learning_rate": 4.9987649590704944e-05,
15
+ "loss": 0.7257,
16
+ "step": 137
17
+ },
18
+ {
19
+ "epoch": 0.020011685655857436,
20
+ "grad_norm": 2.461472988128662,
21
+ "learning_rate": 4.9950610565428546e-05,
22
+ "loss": 0.6506,
23
+ "step": 274
24
+ },
25
+ {
26
+ "epoch": 0.03001752848378615,
27
+ "grad_norm": 2.306278944015503,
28
+ "learning_rate": 4.988891951994058e-05,
29
+ "loss": 0.6267,
30
+ "step": 411
31
+ },
32
+ {
33
+ "epoch": 0.04002337131171487,
34
+ "grad_norm": 2.7277510166168213,
35
+ "learning_rate": 4.9802637407013966e-05,
36
+ "loss": 0.5933,
37
+ "step": 548
38
+ },
39
+ {
40
+ "epoch": 0.05002921413964359,
41
+ "grad_norm": 2.524292230606079,
42
+ "learning_rate": 4.969184947620146e-05,
43
+ "loss": 0.5853,
44
+ "step": 685
45
+ },
46
+ {
47
+ "epoch": 0.0600350569675723,
48
+ "grad_norm": 2.5938446521759033,
49
+ "learning_rate": 4.9556665189606316e-05,
50
+ "loss": 0.5486,
51
+ "step": 822
52
+ },
53
+ {
54
+ "epoch": 0.07004089979550102,
55
+ "grad_norm": 2.528536796569824,
56
+ "learning_rate": 4.93972181137301e-05,
57
+ "loss": 0.5527,
58
+ "step": 959
59
+ },
60
+ {
61
+ "epoch": 0.08004674262342974,
62
+ "grad_norm": 2.1125128269195557,
63
+ "learning_rate": 4.9213665787504655e-05,
64
+ "loss": 0.5129,
65
+ "step": 1096
66
+ },
67
+ {
68
+ "epoch": 0.09005258545135846,
69
+ "grad_norm": 2.1105308532714844,
70
+ "learning_rate": 4.900618956663845e-05,
71
+ "loss": 0.5193,
72
+ "step": 1233
73
+ },
74
+ {
75
+ "epoch": 0.10005842827928718,
76
+ "grad_norm": 1.8278461694717407,
77
+ "learning_rate": 4.877499444443122e-05,
78
+ "loss": 0.5207,
79
+ "step": 1370
80
+ },
81
+ {
82
+ "epoch": 0.1100642711072159,
83
+ "grad_norm": 4.437809944152832,
84
+ "learning_rate": 4.852030884923388e-05,
85
+ "loss": 0.4967,
86
+ "step": 1507
87
+ },
88
+ {
89
+ "epoch": 0.1200701139351446,
90
+ "grad_norm": 2.0539824962615967,
91
+ "learning_rate": 4.82423844187538e-05,
92
+ "loss": 0.4825,
93
+ "step": 1644
94
+ },
95
+ {
96
+ "epoch": 0.13007595676307332,
97
+ "grad_norm": 1.672153115272522,
98
+ "learning_rate": 4.7941495751428536e-05,
99
+ "loss": 0.4648,
100
+ "step": 1781
101
+ },
102
+ {
103
+ "epoch": 0.14008179959100203,
104
+ "grad_norm": 1.9463084936141968,
105
+ "learning_rate": 4.7617940135113606e-05,
106
+ "loss": 0.4623,
107
+ "step": 1918
108
+ },
109
+ {
110
+ "epoch": 0.15008764241893077,
111
+ "grad_norm": 1.61201810836792,
112
+ "learning_rate": 4.7272037253352276e-05,
113
+ "loss": 0.4406,
114
+ "step": 2055
115
+ },
116
+ {
117
+ "epoch": 0.16009348524685948,
118
+ "grad_norm": 1.998997449874878,
119
+ "learning_rate": 4.690412886951786e-05,
120
+ "loss": 0.4563,
121
+ "step": 2192
122
+ },
123
+ {
124
+ "epoch": 0.1700993280747882,
125
+ "grad_norm": 1.4533461332321167,
126
+ "learning_rate": 4.651457848914021e-05,
127
+ "loss": 0.4449,
128
+ "step": 2329
129
+ },
130
+ {
131
+ "epoch": 0.1801051709027169,
132
+ "grad_norm": 1.3883270025253296,
133
+ "learning_rate": 4.610377100075045e-05,
134
+ "loss": 0.4186,
135
+ "step": 2466
136
+ },
137
+ {
138
+ "epoch": 0.19011101373064562,
139
+ "grad_norm": 1.6303733587265015,
140
+ "learning_rate": 4.5672112295598404e-05,
141
+ "loss": 0.42,
142
+ "step": 2603
143
+ },
144
+ {
145
+ "epoch": 0.20011685655857436,
146
+ "grad_norm": 2.2687270641326904,
147
+ "learning_rate": 4.5220028866618837e-05,
148
+ "loss": 0.4063,
149
+ "step": 2740
150
+ },
151
+ {
152
+ "epoch": 0.21012269938650308,
153
+ "grad_norm": 1.6023283004760742,
154
+ "learning_rate": 4.4747967387042424e-05,
155
+ "loss": 0.4037,
156
+ "step": 2877
157
+ },
158
+ {
159
+ "epoch": 0.2201285422144318,
160
+ "grad_norm": 1.6843392848968506,
161
+ "learning_rate": 4.4256394269067967e-05,
162
+ "loss": 0.3942,
163
+ "step": 3014
164
+ },
165
+ {
166
+ "epoch": 0.2301343850423605,
167
+ "grad_norm": 1.3607641458511353,
168
+ "learning_rate": 4.3745795203031904e-05,
169
+ "loss": 0.3898,
170
+ "step": 3151
171
+ },
172
+ {
173
+ "epoch": 0.2401402278702892,
174
+ "grad_norm": 1.3324497938156128,
175
+ "learning_rate": 4.321667467753034e-05,
176
+ "loss": 0.3937,
177
+ "step": 3288
178
+ },
179
+ {
180
+ "epoch": 0.25,
181
+ "eval_loss": 0.4498611092567444,
182
+ "eval_runtime": 365.3631,
183
+ "eval_samples_per_second": 31.58,
184
+ "eval_steps_per_second": 7.896,
185
+ "step": 3423
186
+ },
187
+ {
188
+ "epoch": 0.2501460706982179,
189
+ "grad_norm": 2.132169723510742,
190
+ "learning_rate": 4.2669555480967783e-05,
191
+ "loss": 0.3771,
192
+ "step": 3425
193
+ },
194
+ {
195
+ "epoch": 0.26015191352614664,
196
+ "grad_norm": 1.718522071838379,
197
+ "learning_rate": 4.210497818502509e-05,
198
+ "loss": 0.3865,
199
+ "step": 3562
200
+ },
201
+ {
202
+ "epoch": 0.27015775635407535,
203
+ "grad_norm": 1.6622902154922485,
204
+ "learning_rate": 4.152350061055695e-05,
205
+ "loss": 0.3621,
206
+ "step": 3699
207
+ },
208
+ {
209
+ "epoch": 0.28016359918200406,
210
+ "grad_norm": 1.836055040359497,
211
+ "learning_rate": 4.092569727644661e-05,
212
+ "loss": 0.3764,
213
+ "step": 3836
214
+ },
215
+ {
216
+ "epoch": 0.29016944200993283,
217
+ "grad_norm": 1.3349945545196533,
218
+ "learning_rate": 4.031215883196239e-05,
219
+ "loss": 0.3546,
220
+ "step": 3973
221
+ },
222
+ {
223
+ "epoch": 0.30017528483786154,
224
+ "grad_norm": 1.3506925106048584,
225
+ "learning_rate": 3.968349147317693e-05,
226
+ "loss": 0.3551,
227
+ "step": 4110
228
+ },
229
+ {
230
+ "epoch": 0.31018112766579026,
231
+ "grad_norm": 1.663672685623169,
232
+ "learning_rate": 3.904031634402552e-05,
233
+ "loss": 0.329,
234
+ "step": 4247
235
+ },
236
+ {
237
+ "epoch": 0.32018697049371897,
238
+ "grad_norm": 1.5420622825622559,
239
+ "learning_rate": 3.838326892259564e-05,
240
+ "loss": 0.3477,
241
+ "step": 4384
242
+ },
243
+ {
244
+ "epoch": 0.3301928133216477,
245
+ "grad_norm": 1.14065682888031,
246
+ "learning_rate": 3.7712998393253786e-05,
247
+ "loss": 0.3311,
248
+ "step": 4521
249
+ },
250
+ {
251
+ "epoch": 0.3401986561495764,
252
+ "grad_norm": 2.2005677223205566,
253
+ "learning_rate": 3.703016700522999e-05,
254
+ "loss": 0.3329,
255
+ "step": 4658
256
+ },
257
+ {
258
+ "epoch": 0.3502044989775051,
259
+ "grad_norm": 1.666883111000061,
260
+ "learning_rate": 3.6335449418293985e-05,
261
+ "loss": 0.3398,
262
+ "step": 4795
263
+ },
264
+ {
265
+ "epoch": 0.3602103418054338,
266
+ "grad_norm": 1.7949641942977905,
267
+ "learning_rate": 3.562953203616925e-05,
268
+ "loss": 0.3155,
269
+ "step": 4932
270
+ },
271
+ {
272
+ "epoch": 0.37021618463336253,
273
+ "grad_norm": 1.5265356302261353,
274
+ "learning_rate": 3.491311232834357e-05,
275
+ "loss": 0.3216,
276
+ "step": 5069
277
+ },
278
+ {
279
+ "epoch": 0.38022202746129125,
280
+ "grad_norm": 1.8440781831741333,
281
+ "learning_rate": 3.418689814094646e-05,
282
+ "loss": 0.3154,
283
+ "step": 5206
284
+ },
285
+ {
286
+ "epoch": 0.39022787028921996,
287
+ "grad_norm": 1.4105604887008667,
288
+ "learning_rate": 3.345160699737394e-05,
289
+ "loss": 0.3141,
290
+ "step": 5343
291
+ },
292
+ {
293
+ "epoch": 0.4002337131171487,
294
+ "grad_norm": 1.1201190948486328,
295
+ "learning_rate": 3.2707965389351925e-05,
296
+ "loss": 0.303,
297
+ "step": 5480
298
+ },
299
+ {
300
+ "epoch": 0.41023955594507744,
301
+ "grad_norm": 1.8358474969863892,
302
+ "learning_rate": 3.195670805913866e-05,
303
+ "loss": 0.3042,
304
+ "step": 5617
305
+ },
306
+ {
307
+ "epoch": 0.42024539877300615,
308
+ "grad_norm": 1.7674305438995361,
309
+ "learning_rate": 3.119857727357527e-05,
310
+ "loss": 0.2908,
311
+ "step": 5754
312
+ },
313
+ {
314
+ "epoch": 0.43025124160093486,
315
+ "grad_norm": 1.841166377067566,
316
+ "learning_rate": 3.0434322090701827e-05,
317
+ "loss": 0.2923,
318
+ "step": 5891
319
+ },
320
+ {
321
+ "epoch": 0.4402570844288636,
322
+ "grad_norm": 1.6789052486419678,
323
+ "learning_rate": 2.9664697619663472e-05,
324
+ "loss": 0.2791,
325
+ "step": 6028
326
+ },
327
+ {
328
+ "epoch": 0.4502629272567923,
329
+ "grad_norm": 1.2760952711105347,
330
+ "learning_rate": 2.8890464274637876e-05,
331
+ "loss": 0.2749,
332
+ "step": 6165
333
+ },
334
+ {
335
+ "epoch": 0.460268770084721,
336
+ "grad_norm": 1.3617960214614868,
337
+ "learning_rate": 2.8112387023521115e-05,
338
+ "loss": 0.2715,
339
+ "step": 6302
340
+ },
341
+ {
342
+ "epoch": 0.4702746129126497,
343
+ "grad_norm": 1.0967957973480225,
344
+ "learning_rate": 2.733123463211434e-05,
345
+ "loss": 0.2748,
346
+ "step": 6439
347
+ },
348
+ {
349
+ "epoch": 0.4802804557405784,
350
+ "grad_norm": 0.9176567196846008,
351
+ "learning_rate": 2.6547778904558018e-05,
352
+ "loss": 0.2747,
353
+ "step": 6576
354
+ },
355
+ {
356
+ "epoch": 0.49028629856850714,
357
+ "grad_norm": 0.9466305375099182,
358
+ "learning_rate": 2.5762793920764124e-05,
359
+ "loss": 0.2679,
360
+ "step": 6713
361
+ },
362
+ {
363
+ "epoch": 0.5,
364
+ "eval_loss": 0.31969591975212097,
365
+ "eval_runtime": 364.3381,
366
+ "eval_samples_per_second": 31.668,
367
+ "eval_steps_per_second": 7.918,
368
+ "step": 6846
369
+ },
370
+ {
371
+ "epoch": 0.5002921413964359,
372
+ "grad_norm": 1.4617935419082642,
373
+ "learning_rate": 2.4977055271599893e-05,
374
+ "loss": 0.2643,
375
+ "step": 6850
376
+ },
377
+ {
378
+ "epoch": 0.5102979842243646,
379
+ "grad_norm": 1.6485743522644043,
380
+ "learning_rate": 2.4191339292578617e-05,
381
+ "loss": 0.2605,
382
+ "step": 6987
383
+ },
384
+ {
385
+ "epoch": 0.5203038270522933,
386
+ "grad_norm": 1.0177462100982666,
387
+ "learning_rate": 2.340642229681474e-05,
388
+ "loss": 0.2597,
389
+ "step": 7124
390
+ },
391
+ {
392
+ "epoch": 0.530309669880222,
393
+ "grad_norm": 1.5512287616729736,
394
+ "learning_rate": 2.262307980800109e-05,
395
+ "loss": 0.2427,
396
+ "step": 7261
397
+ },
398
+ {
399
+ "epoch": 0.5403155127081507,
400
+ "grad_norm": 1.314630150794983,
401
+ "learning_rate": 2.1842085794166068e-05,
402
+ "loss": 0.239,
403
+ "step": 7398
404
+ },
405
+ {
406
+ "epoch": 0.5503213555360794,
407
+ "grad_norm": 1.3187059164047241,
408
+ "learning_rate": 2.1064211902967904e-05,
409
+ "loss": 0.2287,
410
+ "step": 7535
411
+ },
412
+ {
413
+ "epoch": 0.5603271983640081,
414
+ "grad_norm": 1.2602145671844482,
415
+ "learning_rate": 2.02902266992815e-05,
416
+ "loss": 0.2381,
417
+ "step": 7672
418
+ },
419
+ {
420
+ "epoch": 0.570333041191937,
421
+ "grad_norm": 1.5069787502288818,
422
+ "learning_rate": 1.9520894905831154e-05,
423
+ "loss": 0.2324,
424
+ "step": 7809
425
+ },
426
+ {
427
+ "epoch": 0.5803388840198657,
428
+ "grad_norm": 0.8896231651306152,
429
+ "learning_rate": 1.8756976647619502e-05,
430
+ "loss": 0.2292,
431
+ "step": 7946
432
+ },
433
+ {
434
+ "epoch": 0.5903447268477944,
435
+ "grad_norm": 0.9988867044448853,
436
+ "learning_rate": 1.7999226700899093e-05,
437
+ "loss": 0.2203,
438
+ "step": 8083
439
+ },
440
+ {
441
+ "epoch": 0.6003505696757231,
442
+ "grad_norm": 0.9557806253433228,
443
+ "learning_rate": 1.7248393747428747e-05,
444
+ "loss": 0.2149,
445
+ "step": 8220
446
+ },
447
+ {
448
+ "epoch": 0.6103564125036518,
449
+ "grad_norm": 1.0728305578231812,
450
+ "learning_rate": 1.6505219634751472e-05,
451
+ "loss": 0.2218,
452
+ "step": 8357
453
+ },
454
+ {
455
+ "epoch": 0.6203622553315805,
456
+ "grad_norm": 1.2526644468307495,
457
+ "learning_rate": 1.5770438643224794e-05,
458
+ "loss": 0.2266,
459
+ "step": 8494
460
+ },
461
+ {
462
+ "epoch": 0.6303680981595092,
463
+ "grad_norm": 1.0386145114898682,
464
+ "learning_rate": 1.5044776760527727e-05,
465
+ "loss": 0.2163,
466
+ "step": 8631
467
+ },
468
+ {
469
+ "epoch": 0.6403739409874379,
470
+ "grad_norm": 1.0314420461654663,
471
+ "learning_rate": 1.4328950964361143e-05,
472
+ "loss": 0.2114,
473
+ "step": 8768
474
+ },
475
+ {
476
+ "epoch": 0.6503797838153667,
477
+ "grad_norm": 0.5962662100791931,
478
+ "learning_rate": 1.3623668514050391e-05,
479
+ "loss": 0.2049,
480
+ "step": 8905
481
+ },
482
+ {
483
+ "epoch": 0.6603856266432954,
484
+ "grad_norm": 0.6801035404205322,
485
+ "learning_rate": 1.2929626251749854e-05,
486
+ "loss": 0.2085,
487
+ "step": 9042
488
+ },
489
+ {
490
+ "epoch": 0.6703914694712241,
491
+ "grad_norm": 1.8030842542648315,
492
+ "learning_rate": 1.2247509913940128e-05,
493
+ "loss": 0.2045,
494
+ "step": 9179
495
+ },
496
+ {
497
+ "epoch": 0.6803973122991528,
498
+ "grad_norm": 1.7659096717834473,
499
+ "learning_rate": 1.1577993453897933e-05,
500
+ "loss": 0.2069,
501
+ "step": 9316
502
+ },
503
+ {
504
+ "epoch": 0.6904031551270815,
505
+ "grad_norm": 0.6933948397636414,
506
+ "learning_rate": 1.0921738375808166e-05,
507
+ "loss": 0.1892,
508
+ "step": 9453
509
+ },
510
+ {
511
+ "epoch": 0.7004089979550102,
512
+ "grad_norm": 1.0741862058639526,
513
+ "learning_rate": 1.0279393081176122e-05,
514
+ "loss": 0.1893,
515
+ "step": 9590
516
+ },
517
+ {
518
+ "epoch": 0.7104148407829389,
519
+ "grad_norm": 0.31156471371650696,
520
+ "learning_rate": 9.651592228185622e-06,
521
+ "loss": 0.1894,
522
+ "step": 9727
523
+ },
524
+ {
525
+ "epoch": 0.7204206836108676,
526
+ "grad_norm": 1.0299690961837769,
527
+ "learning_rate": 9.038956104635871e-06,
528
+ "loss": 0.1844,
529
+ "step": 9864
530
+ },
531
+ {
532
+ "epoch": 0.7304265264387964,
533
+ "grad_norm": 0.8893802165985107,
534
+ "learning_rate": 8.442090015076842e-06,
535
+ "loss": 0.1952,
536
+ "step": 10001
537
+ },
538
+ {
539
+ "epoch": 0.7404323692667251,
540
+ "grad_norm": 1.3043843507766724,
541
+ "learning_rate": 7.861583682748586e-06,
542
+ "loss": 0.1804,
543
+ "step": 10138
544
+ },
545
+ {
546
+ "epoch": 0.75,
547
+ "eval_loss": 0.23675276339054108,
548
+ "eval_runtime": 365.1866,
549
+ "eval_samples_per_second": 31.595,
550
+ "eval_steps_per_second": 7.9,
551
+ "step": 10269
552
+ },
553
+ {
554
+ "epoch": 0.7504382120946538,
555
+ "grad_norm": 1.0938574075698853,
556
+ "learning_rate": 7.298010666915303e-06,
557
+ "loss": 0.1777,
558
+ "step": 10275
559
+ },
560
+ {
561
+ "epoch": 0.7604440549225825,
562
+ "grad_norm": 0.9465618133544922,
563
+ "learning_rate": 6.751927796170044e-06,
564
+ "loss": 0.1902,
565
+ "step": 10412
566
+ },
567
+ {
568
+ "epoch": 0.7704498977505112,
569
+ "grad_norm": 1.0503722429275513,
570
+ "learning_rate": 6.2238746182698375e-06,
571
+ "loss": 0.161,
572
+ "step": 10549
573
+ },
574
+ {
575
+ "epoch": 0.7804557405784399,
576
+ "grad_norm": 1.2977089881896973,
577
+ "learning_rate": 5.7143728670448095e-06,
578
+ "loss": 0.1817,
579
+ "step": 10686
580
+ },
581
+ {
582
+ "epoch": 0.7904615834063686,
583
+ "grad_norm": 0.7178501486778259,
584
+ "learning_rate": 5.223925946908093e-06,
585
+ "loss": 0.1715,
586
+ "step": 10823
587
+ },
588
+ {
589
+ "epoch": 0.8004674262342975,
590
+ "grad_norm": 1.1066858768463135,
591
+ "learning_rate": 4.7530184354757675e-06,
592
+ "loss": 0.1795,
593
+ "step": 10960
594
+ },
595
+ {
596
+ "epoch": 0.8104732690622262,
597
+ "grad_norm": 0.5346119999885559,
598
+ "learning_rate": 4.30211560478837e-06,
599
+ "loss": 0.1716,
600
+ "step": 11097
601
+ },
602
+ {
603
+ "epoch": 0.8204791118901549,
604
+ "grad_norm": 1.6099894046783447,
605
+ "learning_rate": 3.871662961606784e-06,
606
+ "loss": 0.1758,
607
+ "step": 11234
608
+ },
609
+ {
610
+ "epoch": 0.8304849547180836,
611
+ "grad_norm": 0.7507832050323486,
612
+ "learning_rate": 3.4620858072370504e-06,
613
+ "loss": 0.1706,
614
+ "step": 11371
615
+ },
616
+ {
617
+ "epoch": 0.8404907975460123,
618
+ "grad_norm": 1.2874454259872437,
619
+ "learning_rate": 3.073788817318707e-06,
620
+ "loss": 0.1713,
621
+ "step": 11508
622
+ },
623
+ {
624
+ "epoch": 0.850496640373941,
625
+ "grad_norm": 0.863970935344696,
626
+ "learning_rate": 2.7071556419920514e-06,
627
+ "loss": 0.1637,
628
+ "step": 11645
629
+ },
630
+ {
631
+ "epoch": 0.8605024832018697,
632
+ "grad_norm": 0.841557502746582,
633
+ "learning_rate": 2.3625485268391893e-06,
634
+ "loss": 0.1544,
635
+ "step": 11782
636
+ },
637
+ {
638
+ "epoch": 0.8705083260297984,
639
+ "grad_norm": 0.9113953113555908,
640
+ "learning_rate": 2.040307954973572e-06,
641
+ "loss": 0.1642,
642
+ "step": 11919
643
+ },
644
+ {
645
+ "epoch": 0.8805141688577272,
646
+ "grad_norm": 1.1008964776992798,
647
+ "learning_rate": 1.7407523106315244e-06,
648
+ "loss": 0.1501,
649
+ "step": 12056
650
+ },
651
+ {
652
+ "epoch": 0.8905200116856559,
653
+ "grad_norm": 0.7368434071540833,
654
+ "learning_rate": 1.4641775645981849e-06,
655
+ "loss": 0.1617,
656
+ "step": 12193
657
+ },
658
+ {
659
+ "epoch": 0.9005258545135846,
660
+ "grad_norm": 0.8009047508239746,
661
+ "learning_rate": 1.210856981778688e-06,
662
+ "loss": 0.1498,
663
+ "step": 12330
664
+ },
665
+ {
666
+ "epoch": 0.9105316973415133,
667
+ "grad_norm": 0.7741762399673462,
668
+ "learning_rate": 9.810408512034908e-07,
669
+ "loss": 0.1567,
670
+ "step": 12467
671
+ },
672
+ {
673
+ "epoch": 0.920537540169442,
674
+ "grad_norm": 0.5345299243927002,
675
+ "learning_rate": 7.749562387346088e-07,
676
+ "loss": 0.1668,
677
+ "step": 12604
678
+ },
679
+ {
680
+ "epoch": 0.9305433829973707,
681
+ "grad_norm": 0.8907518982887268,
682
+ "learning_rate": 5.928067627171158e-07,
683
+ "loss": 0.159,
684
+ "step": 12741
685
+ },
686
+ {
687
+ "epoch": 0.9405492258252994,
688
+ "grad_norm": 1.0891618728637695,
689
+ "learning_rate": 4.347723927975417e-07,
690
+ "loss": 0.1615,
691
+ "step": 12878
692
+ },
693
+ {
694
+ "epoch": 0.9505550686532281,
695
+ "grad_norm": 0.5715863108634949,
696
+ "learning_rate": 3.01009272107991e-07,
697
+ "loss": 0.1568,
698
+ "step": 13015
699
+ },
700
+ {
701
+ "epoch": 0.9605609114811569,
702
+ "grad_norm": 0.8792664408683777,
703
+ "learning_rate": 1.9164956299158322e-07,
704
+ "loss": 0.1656,
705
+ "step": 13152
706
+ },
707
+ {
708
+ "epoch": 0.9705667543090856,
709
+ "grad_norm": 1.4301509857177734,
710
+ "learning_rate": 1.0680131642176183e-07,
711
+ "loss": 0.1528,
712
+ "step": 13289
713
+ },
714
+ {
715
+ "epoch": 0.9805725971370143,
716
+ "grad_norm": 0.6064629554748535,
717
+ "learning_rate": 4.6548365244375446e-08,
718
+ "loss": 0.1628,
719
+ "step": 13426
720
+ },
721
+ {
722
+ "epoch": 0.990578439964943,
723
+ "grad_norm": 0.9063898921012878,
724
+ "learning_rate": 1.0950241348084422e-08,
725
+ "loss": 0.1529,
726
+ "step": 13563
727
+ },
728
+ {
729
+ "epoch": 1.0,
730
+ "eval_loss": 0.21310460567474365,
731
+ "eval_runtime": 364.3533,
732
+ "eval_samples_per_second": 31.667,
733
+ "eval_steps_per_second": 7.918,
734
+ "step": 13692
735
+ },
736
+ {
737
+ "epoch": 1.0,
738
+ "step": 13692,
739
+ "total_flos": 2.405329205216464e+19,
740
+ "train_loss": 0.29627383469211893,
741
+ "train_runtime": 43035.7563,
742
+ "train_samples_per_second": 2.545,
743
+ "train_steps_per_second": 0.318
744
+ }
745
+ ],
746
+ "logging_steps": 137,
747
+ "max_steps": 13692,
748
+ "num_input_tokens_seen": 0,
749
+ "num_train_epochs": 1,
750
+ "save_steps": 3423,
751
+ "stateful_callbacks": {
752
+ "TrainerControl": {
753
+ "args": {
754
+ "should_epoch_stop": false,
755
+ "should_evaluate": false,
756
+ "should_log": false,
757
+ "should_save": true,
758
+ "should_training_stop": true
759
+ },
760
+ "attributes": {}
761
+ }
762
+ },
763
+ "total_flos": 2.405329205216464e+19,
764
+ "train_batch_size": 2,
765
+ "trial_name": null,
766
+ "trial_params": null
767
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78dd70ff86678b2a66fd86e55abe0dc1ebd9e0a812b7d9efc855bc4004e50c55
3
+ size 7288