Hugging Face's logo

DigitalUmuganda
/
NLLB3.3_finetuned

Safetensors
m2m_100
Model card Files
xet
 Community
Kleber commited on
Commit
cc52e77
·
verified ·
1 Parent(s): 6b255f5

Upload folder using huggingface_hub

Browse files
Files changed (11) hide show
  1. config.json +34 -0
  2. generation_config.json +8 -0
  3. model-00001-of-00003.safetensors +3 -0
  4. model-00002-of-00003.safetensors +3 -0
  5. model-00003-of-00003.safetensors +3 -0
  6. model.safetensors.index.json +1020 -0
  7. optimizer.pt +3 -0
  8. rng_state.pth +3 -0
  9. scheduler.pt +3 -0
  10. trainer_state.json +2193 -0
  11. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/kk/checkpoint-400",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "M2M100ForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "d_model": 2048,
11
+ "decoder_attention_heads": 16,
12
+ "decoder_ffn_dim": 8192,
13
+ "decoder_layerdrop": 0,
14
+ "decoder_layers": 24,
15
+ "decoder_start_token_id": 2,
16
+ "dropout": 0.1,
17
+ "encoder_attention_heads": 16,
18
+ "encoder_ffn_dim": 8192,
19
+ "encoder_layerdrop": 0,
20
+ "encoder_layers": 24,
21
+ "eos_token_id": 2,
22
+ "init_std": 0.02,
23
+ "is_encoder_decoder": true,
24
+ "max_length": 200,
25
+ "max_position_embeddings": 1024,
26
+ "model_type": "m2m_100",
27
+ "num_hidden_layers": 24,
28
+ "pad_token_id": 1,
29
+ "scale_embedding": true,
30
+ "torch_dtype": "float32",
31
+ "transformers_version": "4.44.2",
32
+ "use_cache": true,
33
+ "vocab_size": 256206
34
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 2,
4
+ "eos_token_id": 2,
5
+ "max_length": 200,
6
+ "pad_token_id": 1,
7
+ "transformers_version": "4.44.2"
8
+ }
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:002f0fcf8fd415f514888e58d4fcf6155ef0bf7fd85abc23dceb4fe7b58720c8
3
+ size 4986088344
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03eb31a25b5a3cb1aeba317a10cc2ab6911fb2a3b0ad8f96882a3fd9f0b5c053
3
+ size 4985688360
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fbaf29b41fe1877d2cdef4e0b493b1f6b1245a2415431aa9c8614ab31582a38
3
+ size 3407796744
model.safetensors.index.json ADDED
@@ -0,0 +1,1020 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 13379452928
4
+ },
5
+ "weight_map": {
6
+ "model.decoder.layer_norm.bias": "model-00003-of-00003.safetensors",
7
+ "model.decoder.layer_norm.weight": "model-00003-of-00003.safetensors",
8
+ "model.decoder.layers.0.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
9
+ "model.decoder.layers.0.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
10
+ "model.decoder.layers.0.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
11
+ "model.decoder.layers.0.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
12
+ "model.decoder.layers.0.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
13
+ "model.decoder.layers.0.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
14
+ "model.decoder.layers.0.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
15
+ "model.decoder.layers.0.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
16
+ "model.decoder.layers.0.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
17
+ "model.decoder.layers.0.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
18
+ "model.decoder.layers.0.fc1.bias": "model-00002-of-00003.safetensors",
19
+ "model.decoder.layers.0.fc1.weight": "model-00002-of-00003.safetensors",
20
+ "model.decoder.layers.0.fc2.bias": "model-00002-of-00003.safetensors",
21
+ "model.decoder.layers.0.fc2.weight": "model-00002-of-00003.safetensors",
22
+ "model.decoder.layers.0.final_layer_norm.bias": "model-00002-of-00003.safetensors",
23
+ "model.decoder.layers.0.final_layer_norm.weight": "model-00002-of-00003.safetensors",
24
+ "model.decoder.layers.0.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
25
+ "model.decoder.layers.0.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
26
+ "model.decoder.layers.0.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
27
+ "model.decoder.layers.0.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
28
+ "model.decoder.layers.0.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
29
+ "model.decoder.layers.0.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
30
+ "model.decoder.layers.0.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
31
+ "model.decoder.layers.0.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
32
+ "model.decoder.layers.0.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
33
+ "model.decoder.layers.0.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
34
+ "model.decoder.layers.1.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
35
+ "model.decoder.layers.1.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
36
+ "model.decoder.layers.1.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
37
+ "model.decoder.layers.1.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
38
+ "model.decoder.layers.1.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
39
+ "model.decoder.layers.1.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
40
+ "model.decoder.layers.1.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
41
+ "model.decoder.layers.1.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
42
+ "model.decoder.layers.1.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
43
+ "model.decoder.layers.1.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
44
+ "model.decoder.layers.1.fc1.bias": "model-00002-of-00003.safetensors",
45
+ "model.decoder.layers.1.fc1.weight": "model-00002-of-00003.safetensors",
46
+ "model.decoder.layers.1.fc2.bias": "model-00002-of-00003.safetensors",
47
+ "model.decoder.layers.1.fc2.weight": "model-00002-of-00003.safetensors",
48
+ "model.decoder.layers.1.final_layer_norm.bias": "model-00002-of-00003.safetensors",
49
+ "model.decoder.layers.1.final_layer_norm.weight": "model-00002-of-00003.safetensors",
50
+ "model.decoder.layers.1.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
51
+ "model.decoder.layers.1.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
52
+ "model.decoder.layers.1.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
53
+ "model.decoder.layers.1.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
54
+ "model.decoder.layers.1.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
55
+ "model.decoder.layers.1.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
56
+ "model.decoder.layers.1.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
57
+ "model.decoder.layers.1.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
58
+ "model.decoder.layers.1.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
59
+ "model.decoder.layers.1.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
60
+ "model.decoder.layers.10.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
61
+ "model.decoder.layers.10.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
62
+ "model.decoder.layers.10.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
63
+ "model.decoder.layers.10.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
64
+ "model.decoder.layers.10.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
65
+ "model.decoder.layers.10.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
66
+ "model.decoder.layers.10.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
67
+ "model.decoder.layers.10.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
68
+ "model.decoder.layers.10.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
69
+ "model.decoder.layers.10.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
70
+ "model.decoder.layers.10.fc1.bias": "model-00002-of-00003.safetensors",
71
+ "model.decoder.layers.10.fc1.weight": "model-00002-of-00003.safetensors",
72
+ "model.decoder.layers.10.fc2.bias": "model-00002-of-00003.safetensors",
73
+ "model.decoder.layers.10.fc2.weight": "model-00002-of-00003.safetensors",
74
+ "model.decoder.layers.10.final_layer_norm.bias": "model-00002-of-00003.safetensors",
75
+ "model.decoder.layers.10.final_layer_norm.weight": "model-00002-of-00003.safetensors",
76
+ "model.decoder.layers.10.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
77
+ "model.decoder.layers.10.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
78
+ "model.decoder.layers.10.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
79
+ "model.decoder.layers.10.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
80
+ "model.decoder.layers.10.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
81
+ "model.decoder.layers.10.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
82
+ "model.decoder.layers.10.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
83
+ "model.decoder.layers.10.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
84
+ "model.decoder.layers.10.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
85
+ "model.decoder.layers.10.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
86
+ "model.decoder.layers.11.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
87
+ "model.decoder.layers.11.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
88
+ "model.decoder.layers.11.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
89
+ "model.decoder.layers.11.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
90
+ "model.decoder.layers.11.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
91
+ "model.decoder.layers.11.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
92
+ "model.decoder.layers.11.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
93
+ "model.decoder.layers.11.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
94
+ "model.decoder.layers.11.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
95
+ "model.decoder.layers.11.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
96
+ "model.decoder.layers.11.fc1.bias": "model-00003-of-00003.safetensors",
97
+ "model.decoder.layers.11.fc1.weight": "model-00003-of-00003.safetensors",
98
+ "model.decoder.layers.11.fc2.bias": "model-00003-of-00003.safetensors",
99
+ "model.decoder.layers.11.fc2.weight": "model-00003-of-00003.safetensors",
100
+ "model.decoder.layers.11.final_layer_norm.bias": "model-00003-of-00003.safetensors",
101
+ "model.decoder.layers.11.final_layer_norm.weight": "model-00003-of-00003.safetensors",
102
+ "model.decoder.layers.11.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
103
+ "model.decoder.layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
104
+ "model.decoder.layers.11.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
105
+ "model.decoder.layers.11.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
106
+ "model.decoder.layers.11.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
107
+ "model.decoder.layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
108
+ "model.decoder.layers.11.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
109
+ "model.decoder.layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
110
+ "model.decoder.layers.11.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
111
+ "model.decoder.layers.11.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
112
+ "model.decoder.layers.12.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
113
+ "model.decoder.layers.12.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
114
+ "model.decoder.layers.12.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
115
+ "model.decoder.layers.12.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
116
+ "model.decoder.layers.12.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
117
+ "model.decoder.layers.12.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
118
+ "model.decoder.layers.12.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
119
+ "model.decoder.layers.12.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
120
+ "model.decoder.layers.12.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
121
+ "model.decoder.layers.12.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
122
+ "model.decoder.layers.12.fc1.bias": "model-00003-of-00003.safetensors",
123
+ "model.decoder.layers.12.fc1.weight": "model-00003-of-00003.safetensors",
124
+ "model.decoder.layers.12.fc2.bias": "model-00003-of-00003.safetensors",
125
+ "model.decoder.layers.12.fc2.weight": "model-00003-of-00003.safetensors",
126
+ "model.decoder.layers.12.final_layer_norm.bias": "model-00003-of-00003.safetensors",
127
+ "model.decoder.layers.12.final_layer_norm.weight": "model-00003-of-00003.safetensors",
128
+ "model.decoder.layers.12.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
129
+ "model.decoder.layers.12.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
130
+ "model.decoder.layers.12.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
131
+ "model.decoder.layers.12.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
132
+ "model.decoder.layers.12.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
133
+ "model.decoder.layers.12.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
134
+ "model.decoder.layers.12.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
135
+ "model.decoder.layers.12.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
136
+ "model.decoder.layers.12.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
137
+ "model.decoder.layers.12.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
138
+ "model.decoder.layers.13.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
139
+ "model.decoder.layers.13.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
140
+ "model.decoder.layers.13.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
141
+ "model.decoder.layers.13.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
142
+ "model.decoder.layers.13.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
143
+ "model.decoder.layers.13.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
144
+ "model.decoder.layers.13.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
145
+ "model.decoder.layers.13.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
146
+ "model.decoder.layers.13.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
147
+ "model.decoder.layers.13.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
148
+ "model.decoder.layers.13.fc1.bias": "model-00003-of-00003.safetensors",
149
+ "model.decoder.layers.13.fc1.weight": "model-00003-of-00003.safetensors",
150
+ "model.decoder.layers.13.fc2.bias": "model-00003-of-00003.safetensors",
151
+ "model.decoder.layers.13.fc2.weight": "model-00003-of-00003.safetensors",
152
+ "model.decoder.layers.13.final_layer_norm.bias": "model-00003-of-00003.safetensors",
153
+ "model.decoder.layers.13.final_layer_norm.weight": "model-00003-of-00003.safetensors",
154
+ "model.decoder.layers.13.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
155
+ "model.decoder.layers.13.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
156
+ "model.decoder.layers.13.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
157
+ "model.decoder.layers.13.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
158
+ "model.decoder.layers.13.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
159
+ "model.decoder.layers.13.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
160
+ "model.decoder.layers.13.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
161
+ "model.decoder.layers.13.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
162
+ "model.decoder.layers.13.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
163
+ "model.decoder.layers.13.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
164
+ "model.decoder.layers.14.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
165
+ "model.decoder.layers.14.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
166
+ "model.decoder.layers.14.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
167
+ "model.decoder.layers.14.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
168
+ "model.decoder.layers.14.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
169
+ "model.decoder.layers.14.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
170
+ "model.decoder.layers.14.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
171
+ "model.decoder.layers.14.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
172
+ "model.decoder.layers.14.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
173
+ "model.decoder.layers.14.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
174
+ "model.decoder.layers.14.fc1.bias": "model-00003-of-00003.safetensors",
175
+ "model.decoder.layers.14.fc1.weight": "model-00003-of-00003.safetensors",
176
+ "model.decoder.layers.14.fc2.bias": "model-00003-of-00003.safetensors",
177
+ "model.decoder.layers.14.fc2.weight": "model-00003-of-00003.safetensors",
178
+ "model.decoder.layers.14.final_layer_norm.bias": "model-00003-of-00003.safetensors",
179
+ "model.decoder.layers.14.final_layer_norm.weight": "model-00003-of-00003.safetensors",
180
+ "model.decoder.layers.14.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
181
+ "model.decoder.layers.14.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
182
+ "model.decoder.layers.14.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
183
+ "model.decoder.layers.14.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
184
+ "model.decoder.layers.14.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
185
+ "model.decoder.layers.14.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
186
+ "model.decoder.layers.14.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
187
+ "model.decoder.layers.14.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
188
+ "model.decoder.layers.14.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
189
+ "model.decoder.layers.14.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
190
+ "model.decoder.layers.15.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
191
+ "model.decoder.layers.15.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
192
+ "model.decoder.layers.15.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
193
+ "model.decoder.layers.15.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
194
+ "model.decoder.layers.15.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
195
+ "model.decoder.layers.15.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
196
+ "model.decoder.layers.15.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
197
+ "model.decoder.layers.15.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
198
+ "model.decoder.layers.15.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
199
+ "model.decoder.layers.15.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
200
+ "model.decoder.layers.15.fc1.bias": "model-00003-of-00003.safetensors",
201
+ "model.decoder.layers.15.fc1.weight": "model-00003-of-00003.safetensors",
202
+ "model.decoder.layers.15.fc2.bias": "model-00003-of-00003.safetensors",
203
+ "model.decoder.layers.15.fc2.weight": "model-00003-of-00003.safetensors",
204
+ "model.decoder.layers.15.final_layer_norm.bias": "model-00003-of-00003.safetensors",
205
+ "model.decoder.layers.15.final_layer_norm.weight": "model-00003-of-00003.safetensors",
206
+ "model.decoder.layers.15.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
207
+ "model.decoder.layers.15.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
208
+ "model.decoder.layers.15.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
209
+ "model.decoder.layers.15.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
210
+ "model.decoder.layers.15.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
211
+ "model.decoder.layers.15.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
212
+ "model.decoder.layers.15.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
213
+ "model.decoder.layers.15.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
214
+ "model.decoder.layers.15.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
215
+ "model.decoder.layers.15.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
216
+ "model.decoder.layers.16.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
217
+ "model.decoder.layers.16.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
218
+ "model.decoder.layers.16.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
219
+ "model.decoder.layers.16.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
220
+ "model.decoder.layers.16.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
221
+ "model.decoder.layers.16.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
222
+ "model.decoder.layers.16.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
223
+ "model.decoder.layers.16.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
224
+ "model.decoder.layers.16.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
225
+ "model.decoder.layers.16.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
226
+ "model.decoder.layers.16.fc1.bias": "model-00003-of-00003.safetensors",
227
+ "model.decoder.layers.16.fc1.weight": "model-00003-of-00003.safetensors",
228
+ "model.decoder.layers.16.fc2.bias": "model-00003-of-00003.safetensors",
229
+ "model.decoder.layers.16.fc2.weight": "model-00003-of-00003.safetensors",
230
+ "model.decoder.layers.16.final_layer_norm.bias": "model-00003-of-00003.safetensors",
231
+ "model.decoder.layers.16.final_layer_norm.weight": "model-00003-of-00003.safetensors",
232
+ "model.decoder.layers.16.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
233
+ "model.decoder.layers.16.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
234
+ "model.decoder.layers.16.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
235
+ "model.decoder.layers.16.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
236
+ "model.decoder.layers.16.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
237
+ "model.decoder.layers.16.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
238
+ "model.decoder.layers.16.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
239
+ "model.decoder.layers.16.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
240
+ "model.decoder.layers.16.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
241
+ "model.decoder.layers.16.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
242
+ "model.decoder.layers.17.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
243
+ "model.decoder.layers.17.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
244
+ "model.decoder.layers.17.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
245
+ "model.decoder.layers.17.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
246
+ "model.decoder.layers.17.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
247
+ "model.decoder.layers.17.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
248
+ "model.decoder.layers.17.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
249
+ "model.decoder.layers.17.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
250
+ "model.decoder.layers.17.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
251
+ "model.decoder.layers.17.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
252
+ "model.decoder.layers.17.fc1.bias": "model-00003-of-00003.safetensors",
253
+ "model.decoder.layers.17.fc1.weight": "model-00003-of-00003.safetensors",
254
+ "model.decoder.layers.17.fc2.bias": "model-00003-of-00003.safetensors",
255
+ "model.decoder.layers.17.fc2.weight": "model-00003-of-00003.safetensors",
256
+ "model.decoder.layers.17.final_layer_norm.bias": "model-00003-of-00003.safetensors",
257
+ "model.decoder.layers.17.final_layer_norm.weight": "model-00003-of-00003.safetensors",
258
+ "model.decoder.layers.17.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
259
+ "model.decoder.layers.17.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
260
+ "model.decoder.layers.17.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
261
+ "model.decoder.layers.17.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
262
+ "model.decoder.layers.17.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
263
+ "model.decoder.layers.17.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
264
+ "model.decoder.layers.17.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
265
+ "model.decoder.layers.17.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
266
+ "model.decoder.layers.17.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
267
+ "model.decoder.layers.17.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
268
+ "model.decoder.layers.18.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
269
+ "model.decoder.layers.18.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
270
+ "model.decoder.layers.18.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
271
+ "model.decoder.layers.18.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
272
+ "model.decoder.layers.18.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
273
+ "model.decoder.layers.18.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
274
+ "model.decoder.layers.18.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
275
+ "model.decoder.layers.18.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
276
+ "model.decoder.layers.18.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
277
+ "model.decoder.layers.18.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
278
+ "model.decoder.layers.18.fc1.bias": "model-00003-of-00003.safetensors",
279
+ "model.decoder.layers.18.fc1.weight": "model-00003-of-00003.safetensors",
280
+ "model.decoder.layers.18.fc2.bias": "model-00003-of-00003.safetensors",
281
+ "model.decoder.layers.18.fc2.weight": "model-00003-of-00003.safetensors",
282
+ "model.decoder.layers.18.final_layer_norm.bias": "model-00003-of-00003.safetensors",
283
+ "model.decoder.layers.18.final_layer_norm.weight": "model-00003-of-00003.safetensors",
284
+ "model.decoder.layers.18.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
285
+ "model.decoder.layers.18.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
286
+ "model.decoder.layers.18.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
287
+ "model.decoder.layers.18.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
288
+ "model.decoder.layers.18.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
289
+ "model.decoder.layers.18.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
290
+ "model.decoder.layers.18.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
291
+ "model.decoder.layers.18.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
292
+ "model.decoder.layers.18.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
293
+ "model.decoder.layers.18.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
294
+ "model.decoder.layers.19.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
295
+ "model.decoder.layers.19.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
296
+ "model.decoder.layers.19.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
297
+ "model.decoder.layers.19.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
298
+ "model.decoder.layers.19.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
299
+ "model.decoder.layers.19.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
300
+ "model.decoder.layers.19.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
301
+ "model.decoder.layers.19.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
302
+ "model.decoder.layers.19.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
303
+ "model.decoder.layers.19.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
304
+ "model.decoder.layers.19.fc1.bias": "model-00003-of-00003.safetensors",
305
+ "model.decoder.layers.19.fc1.weight": "model-00003-of-00003.safetensors",
306
+ "model.decoder.layers.19.fc2.bias": "model-00003-of-00003.safetensors",
307
+ "model.decoder.layers.19.fc2.weight": "model-00003-of-00003.safetensors",
308
+ "model.decoder.layers.19.final_layer_norm.bias": "model-00003-of-00003.safetensors",
309
+ "model.decoder.layers.19.final_layer_norm.weight": "model-00003-of-00003.safetensors",
310
+ "model.decoder.layers.19.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
311
+ "model.decoder.layers.19.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
312
+ "model.decoder.layers.19.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
313
+ "model.decoder.layers.19.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
314
+ "model.decoder.layers.19.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
315
+ "model.decoder.layers.19.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
316
+ "model.decoder.layers.19.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
317
+ "model.decoder.layers.19.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
318
+ "model.decoder.layers.19.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
319
+ "model.decoder.layers.19.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
320
+ "model.decoder.layers.2.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
321
+ "model.decoder.layers.2.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
322
+ "model.decoder.layers.2.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
323
+ "model.decoder.layers.2.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
324
+ "model.decoder.layers.2.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
325
+ "model.decoder.layers.2.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
326
+ "model.decoder.layers.2.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
327
+ "model.decoder.layers.2.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
328
+ "model.decoder.layers.2.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
329
+ "model.decoder.layers.2.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
330
+ "model.decoder.layers.2.fc1.bias": "model-00002-of-00003.safetensors",
331
+ "model.decoder.layers.2.fc1.weight": "model-00002-of-00003.safetensors",
332
+ "model.decoder.layers.2.fc2.bias": "model-00002-of-00003.safetensors",
333
+ "model.decoder.layers.2.fc2.weight": "model-00002-of-00003.safetensors",
334
+ "model.decoder.layers.2.final_layer_norm.bias": "model-00002-of-00003.safetensors",
335
+ "model.decoder.layers.2.final_layer_norm.weight": "model-00002-of-00003.safetensors",
336
+ "model.decoder.layers.2.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
337
+ "model.decoder.layers.2.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
338
+ "model.decoder.layers.2.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
339
+ "model.decoder.layers.2.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
340
+ "model.decoder.layers.2.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
341
+ "model.decoder.layers.2.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
342
+ "model.decoder.layers.2.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
343
+ "model.decoder.layers.2.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
344
+ "model.decoder.layers.2.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
345
+ "model.decoder.layers.2.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
346
+ "model.decoder.layers.20.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
347
+ "model.decoder.layers.20.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
348
+ "model.decoder.layers.20.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
349
+ "model.decoder.layers.20.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
350
+ "model.decoder.layers.20.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
351
+ "model.decoder.layers.20.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
352
+ "model.decoder.layers.20.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
353
+ "model.decoder.layers.20.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
354
+ "model.decoder.layers.20.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
355
+ "model.decoder.layers.20.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
356
+ "model.decoder.layers.20.fc1.bias": "model-00003-of-00003.safetensors",
357
+ "model.decoder.layers.20.fc1.weight": "model-00003-of-00003.safetensors",
358
+ "model.decoder.layers.20.fc2.bias": "model-00003-of-00003.safetensors",
359
+ "model.decoder.layers.20.fc2.weight": "model-00003-of-00003.safetensors",
360
+ "model.decoder.layers.20.final_layer_norm.bias": "model-00003-of-00003.safetensors",
361
+ "model.decoder.layers.20.final_layer_norm.weight": "model-00003-of-00003.safetensors",
362
+ "model.decoder.layers.20.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
363
+ "model.decoder.layers.20.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
364
+ "model.decoder.layers.20.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
365
+ "model.decoder.layers.20.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
366
+ "model.decoder.layers.20.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
367
+ "model.decoder.layers.20.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
368
+ "model.decoder.layers.20.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
369
+ "model.decoder.layers.20.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
370
+ "model.decoder.layers.20.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
371
+ "model.decoder.layers.20.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
372
+ "model.decoder.layers.21.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
373
+ "model.decoder.layers.21.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
374
+ "model.decoder.layers.21.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
375
+ "model.decoder.layers.21.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
376
+ "model.decoder.layers.21.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
377
+ "model.decoder.layers.21.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
378
+ "model.decoder.layers.21.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
379
+ "model.decoder.layers.21.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
380
+ "model.decoder.layers.21.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
381
+ "model.decoder.layers.21.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
382
+ "model.decoder.layers.21.fc1.bias": "model-00003-of-00003.safetensors",
383
+ "model.decoder.layers.21.fc1.weight": "model-00003-of-00003.safetensors",
384
+ "model.decoder.layers.21.fc2.bias": "model-00003-of-00003.safetensors",
385
+ "model.decoder.layers.21.fc2.weight": "model-00003-of-00003.safetensors",
386
+ "model.decoder.layers.21.final_layer_norm.bias": "model-00003-of-00003.safetensors",
387
+ "model.decoder.layers.21.final_layer_norm.weight": "model-00003-of-00003.safetensors",
388
+ "model.decoder.layers.21.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
389
+ "model.decoder.layers.21.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
390
+ "model.decoder.layers.21.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
391
+ "model.decoder.layers.21.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
392
+ "model.decoder.layers.21.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
393
+ "model.decoder.layers.21.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
394
+ "model.decoder.layers.21.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
395
+ "model.decoder.layers.21.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
396
+ "model.decoder.layers.21.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
397
+ "model.decoder.layers.21.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
398
+ "model.decoder.layers.22.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
399
+ "model.decoder.layers.22.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
400
+ "model.decoder.layers.22.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
401
+ "model.decoder.layers.22.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
402
+ "model.decoder.layers.22.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
403
+ "model.decoder.layers.22.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
404
+ "model.decoder.layers.22.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
405
+ "model.decoder.layers.22.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
406
+ "model.decoder.layers.22.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
407
+ "model.decoder.layers.22.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
408
+ "model.decoder.layers.22.fc1.bias": "model-00003-of-00003.safetensors",
409
+ "model.decoder.layers.22.fc1.weight": "model-00003-of-00003.safetensors",
410
+ "model.decoder.layers.22.fc2.bias": "model-00003-of-00003.safetensors",
411
+ "model.decoder.layers.22.fc2.weight": "model-00003-of-00003.safetensors",
412
+ "model.decoder.layers.22.final_layer_norm.bias": "model-00003-of-00003.safetensors",
413
+ "model.decoder.layers.22.final_layer_norm.weight": "model-00003-of-00003.safetensors",
414
+ "model.decoder.layers.22.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
415
+ "model.decoder.layers.22.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
416
+ "model.decoder.layers.22.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
417
+ "model.decoder.layers.22.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
418
+ "model.decoder.layers.22.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
419
+ "model.decoder.layers.22.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
420
+ "model.decoder.layers.22.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
421
+ "model.decoder.layers.22.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
422
+ "model.decoder.layers.22.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
423
+ "model.decoder.layers.22.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
424
+ "model.decoder.layers.23.encoder_attn.k_proj.bias": "model-00003-of-00003.safetensors",
425
+ "model.decoder.layers.23.encoder_attn.k_proj.weight": "model-00003-of-00003.safetensors",
426
+ "model.decoder.layers.23.encoder_attn.out_proj.bias": "model-00003-of-00003.safetensors",
427
+ "model.decoder.layers.23.encoder_attn.out_proj.weight": "model-00003-of-00003.safetensors",
428
+ "model.decoder.layers.23.encoder_attn.q_proj.bias": "model-00003-of-00003.safetensors",
429
+ "model.decoder.layers.23.encoder_attn.q_proj.weight": "model-00003-of-00003.safetensors",
430
+ "model.decoder.layers.23.encoder_attn.v_proj.bias": "model-00003-of-00003.safetensors",
431
+ "model.decoder.layers.23.encoder_attn.v_proj.weight": "model-00003-of-00003.safetensors",
432
+ "model.decoder.layers.23.encoder_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
433
+ "model.decoder.layers.23.encoder_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
434
+ "model.decoder.layers.23.fc1.bias": "model-00003-of-00003.safetensors",
435
+ "model.decoder.layers.23.fc1.weight": "model-00003-of-00003.safetensors",
436
+ "model.decoder.layers.23.fc2.bias": "model-00003-of-00003.safetensors",
437
+ "model.decoder.layers.23.fc2.weight": "model-00003-of-00003.safetensors",
438
+ "model.decoder.layers.23.final_layer_norm.bias": "model-00003-of-00003.safetensors",
439
+ "model.decoder.layers.23.final_layer_norm.weight": "model-00003-of-00003.safetensors",
440
+ "model.decoder.layers.23.self_attn.k_proj.bias": "model-00003-of-00003.safetensors",
441
+ "model.decoder.layers.23.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
442
+ "model.decoder.layers.23.self_attn.out_proj.bias": "model-00003-of-00003.safetensors",
443
+ "model.decoder.layers.23.self_attn.out_proj.weight": "model-00003-of-00003.safetensors",
444
+ "model.decoder.layers.23.self_attn.q_proj.bias": "model-00003-of-00003.safetensors",
445
+ "model.decoder.layers.23.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
446
+ "model.decoder.layers.23.self_attn.v_proj.bias": "model-00003-of-00003.safetensors",
447
+ "model.decoder.layers.23.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
448
+ "model.decoder.layers.23.self_attn_layer_norm.bias": "model-00003-of-00003.safetensors",
449
+ "model.decoder.layers.23.self_attn_layer_norm.weight": "model-00003-of-00003.safetensors",
450
+ "model.decoder.layers.3.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
451
+ "model.decoder.layers.3.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
452
+ "model.decoder.layers.3.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
453
+ "model.decoder.layers.3.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
454
+ "model.decoder.layers.3.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
455
+ "model.decoder.layers.3.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
456
+ "model.decoder.layers.3.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
457
+ "model.decoder.layers.3.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
458
+ "model.decoder.layers.3.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
459
+ "model.decoder.layers.3.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
460
+ "model.decoder.layers.3.fc1.bias": "model-00002-of-00003.safetensors",
461
+ "model.decoder.layers.3.fc1.weight": "model-00002-of-00003.safetensors",
462
+ "model.decoder.layers.3.fc2.bias": "model-00002-of-00003.safetensors",
463
+ "model.decoder.layers.3.fc2.weight": "model-00002-of-00003.safetensors",
464
+ "model.decoder.layers.3.final_layer_norm.bias": "model-00002-of-00003.safetensors",
465
+ "model.decoder.layers.3.final_layer_norm.weight": "model-00002-of-00003.safetensors",
466
+ "model.decoder.layers.3.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
467
+ "model.decoder.layers.3.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
468
+ "model.decoder.layers.3.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
469
+ "model.decoder.layers.3.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
470
+ "model.decoder.layers.3.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
471
+ "model.decoder.layers.3.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
472
+ "model.decoder.layers.3.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
473
+ "model.decoder.layers.3.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
474
+ "model.decoder.layers.3.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
475
+ "model.decoder.layers.3.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
476
+ "model.decoder.layers.4.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
477
+ "model.decoder.layers.4.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
478
+ "model.decoder.layers.4.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
479
+ "model.decoder.layers.4.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
480
+ "model.decoder.layers.4.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
481
+ "model.decoder.layers.4.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
482
+ "model.decoder.layers.4.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
483
+ "model.decoder.layers.4.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
484
+ "model.decoder.layers.4.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
485
+ "model.decoder.layers.4.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
486
+ "model.decoder.layers.4.fc1.bias": "model-00002-of-00003.safetensors",
487
+ "model.decoder.layers.4.fc1.weight": "model-00002-of-00003.safetensors",
488
+ "model.decoder.layers.4.fc2.bias": "model-00002-of-00003.safetensors",
489
+ "model.decoder.layers.4.fc2.weight": "model-00002-of-00003.safetensors",
490
+ "model.decoder.layers.4.final_layer_norm.bias": "model-00002-of-00003.safetensors",
491
+ "model.decoder.layers.4.final_layer_norm.weight": "model-00002-of-00003.safetensors",
492
+ "model.decoder.layers.4.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
493
+ "model.decoder.layers.4.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
494
+ "model.decoder.layers.4.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
495
+ "model.decoder.layers.4.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
496
+ "model.decoder.layers.4.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
497
+ "model.decoder.layers.4.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
498
+ "model.decoder.layers.4.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
499
+ "model.decoder.layers.4.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
500
+ "model.decoder.layers.4.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
501
+ "model.decoder.layers.4.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
502
+ "model.decoder.layers.5.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
503
+ "model.decoder.layers.5.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
504
+ "model.decoder.layers.5.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
505
+ "model.decoder.layers.5.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
506
+ "model.decoder.layers.5.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
507
+ "model.decoder.layers.5.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
508
+ "model.decoder.layers.5.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
509
+ "model.decoder.layers.5.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
510
+ "model.decoder.layers.5.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
511
+ "model.decoder.layers.5.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
512
+ "model.decoder.layers.5.fc1.bias": "model-00002-of-00003.safetensors",
513
+ "model.decoder.layers.5.fc1.weight": "model-00002-of-00003.safetensors",
514
+ "model.decoder.layers.5.fc2.bias": "model-00002-of-00003.safetensors",
515
+ "model.decoder.layers.5.fc2.weight": "model-00002-of-00003.safetensors",
516
+ "model.decoder.layers.5.final_layer_norm.bias": "model-00002-of-00003.safetensors",
517
+ "model.decoder.layers.5.final_layer_norm.weight": "model-00002-of-00003.safetensors",
518
+ "model.decoder.layers.5.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
519
+ "model.decoder.layers.5.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
520
+ "model.decoder.layers.5.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
521
+ "model.decoder.layers.5.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
522
+ "model.decoder.layers.5.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
523
+ "model.decoder.layers.5.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
524
+ "model.decoder.layers.5.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
525
+ "model.decoder.layers.5.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
526
+ "model.decoder.layers.5.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
527
+ "model.decoder.layers.5.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
528
+ "model.decoder.layers.6.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
529
+ "model.decoder.layers.6.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
530
+ "model.decoder.layers.6.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
531
+ "model.decoder.layers.6.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
532
+ "model.decoder.layers.6.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
533
+ "model.decoder.layers.6.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
534
+ "model.decoder.layers.6.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
535
+ "model.decoder.layers.6.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
536
+ "model.decoder.layers.6.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
537
+ "model.decoder.layers.6.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
538
+ "model.decoder.layers.6.fc1.bias": "model-00002-of-00003.safetensors",
539
+ "model.decoder.layers.6.fc1.weight": "model-00002-of-00003.safetensors",
540
+ "model.decoder.layers.6.fc2.bias": "model-00002-of-00003.safetensors",
541
+ "model.decoder.layers.6.fc2.weight": "model-00002-of-00003.safetensors",
542
+ "model.decoder.layers.6.final_layer_norm.bias": "model-00002-of-00003.safetensors",
543
+ "model.decoder.layers.6.final_layer_norm.weight": "model-00002-of-00003.safetensors",
544
+ "model.decoder.layers.6.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
545
+ "model.decoder.layers.6.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
546
+ "model.decoder.layers.6.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
547
+ "model.decoder.layers.6.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
548
+ "model.decoder.layers.6.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
549
+ "model.decoder.layers.6.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
550
+ "model.decoder.layers.6.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
551
+ "model.decoder.layers.6.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
552
+ "model.decoder.layers.6.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
553
+ "model.decoder.layers.6.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
554
+ "model.decoder.layers.7.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
555
+ "model.decoder.layers.7.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
556
+ "model.decoder.layers.7.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
557
+ "model.decoder.layers.7.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
558
+ "model.decoder.layers.7.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
559
+ "model.decoder.layers.7.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
560
+ "model.decoder.layers.7.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
561
+ "model.decoder.layers.7.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
562
+ "model.decoder.layers.7.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
563
+ "model.decoder.layers.7.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
564
+ "model.decoder.layers.7.fc1.bias": "model-00002-of-00003.safetensors",
565
+ "model.decoder.layers.7.fc1.weight": "model-00002-of-00003.safetensors",
566
+ "model.decoder.layers.7.fc2.bias": "model-00002-of-00003.safetensors",
567
+ "model.decoder.layers.7.fc2.weight": "model-00002-of-00003.safetensors",
568
+ "model.decoder.layers.7.final_layer_norm.bias": "model-00002-of-00003.safetensors",
569
+ "model.decoder.layers.7.final_layer_norm.weight": "model-00002-of-00003.safetensors",
570
+ "model.decoder.layers.7.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
571
+ "model.decoder.layers.7.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
572
+ "model.decoder.layers.7.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
573
+ "model.decoder.layers.7.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
574
+ "model.decoder.layers.7.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
575
+ "model.decoder.layers.7.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
576
+ "model.decoder.layers.7.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
577
+ "model.decoder.layers.7.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
578
+ "model.decoder.layers.7.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
579
+ "model.decoder.layers.7.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
580
+ "model.decoder.layers.8.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
581
+ "model.decoder.layers.8.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
582
+ "model.decoder.layers.8.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
583
+ "model.decoder.layers.8.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
584
+ "model.decoder.layers.8.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
585
+ "model.decoder.layers.8.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
586
+ "model.decoder.layers.8.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
587
+ "model.decoder.layers.8.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
588
+ "model.decoder.layers.8.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
589
+ "model.decoder.layers.8.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
590
+ "model.decoder.layers.8.fc1.bias": "model-00002-of-00003.safetensors",
591
+ "model.decoder.layers.8.fc1.weight": "model-00002-of-00003.safetensors",
592
+ "model.decoder.layers.8.fc2.bias": "model-00002-of-00003.safetensors",
593
+ "model.decoder.layers.8.fc2.weight": "model-00002-of-00003.safetensors",
594
+ "model.decoder.layers.8.final_layer_norm.bias": "model-00002-of-00003.safetensors",
595
+ "model.decoder.layers.8.final_layer_norm.weight": "model-00002-of-00003.safetensors",
596
+ "model.decoder.layers.8.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
597
+ "model.decoder.layers.8.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
598
+ "model.decoder.layers.8.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
599
+ "model.decoder.layers.8.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
600
+ "model.decoder.layers.8.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
601
+ "model.decoder.layers.8.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
602
+ "model.decoder.layers.8.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
603
+ "model.decoder.layers.8.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
604
+ "model.decoder.layers.8.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
605
+ "model.decoder.layers.8.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
606
+ "model.decoder.layers.9.encoder_attn.k_proj.bias": "model-00002-of-00003.safetensors",
607
+ "model.decoder.layers.9.encoder_attn.k_proj.weight": "model-00002-of-00003.safetensors",
608
+ "model.decoder.layers.9.encoder_attn.out_proj.bias": "model-00002-of-00003.safetensors",
609
+ "model.decoder.layers.9.encoder_attn.out_proj.weight": "model-00002-of-00003.safetensors",
610
+ "model.decoder.layers.9.encoder_attn.q_proj.bias": "model-00002-of-00003.safetensors",
611
+ "model.decoder.layers.9.encoder_attn.q_proj.weight": "model-00002-of-00003.safetensors",
612
+ "model.decoder.layers.9.encoder_attn.v_proj.bias": "model-00002-of-00003.safetensors",
613
+ "model.decoder.layers.9.encoder_attn.v_proj.weight": "model-00002-of-00003.safetensors",
614
+ "model.decoder.layers.9.encoder_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
615
+ "model.decoder.layers.9.encoder_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
616
+ "model.decoder.layers.9.fc1.bias": "model-00002-of-00003.safetensors",
617
+ "model.decoder.layers.9.fc1.weight": "model-00002-of-00003.safetensors",
618
+ "model.decoder.layers.9.fc2.bias": "model-00002-of-00003.safetensors",
619
+ "model.decoder.layers.9.fc2.weight": "model-00002-of-00003.safetensors",
620
+ "model.decoder.layers.9.final_layer_norm.bias": "model-00002-of-00003.safetensors",
621
+ "model.decoder.layers.9.final_layer_norm.weight": "model-00002-of-00003.safetensors",
622
+ "model.decoder.layers.9.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
623
+ "model.decoder.layers.9.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
624
+ "model.decoder.layers.9.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
625
+ "model.decoder.layers.9.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
626
+ "model.decoder.layers.9.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
627
+ "model.decoder.layers.9.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
628
+ "model.decoder.layers.9.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
629
+ "model.decoder.layers.9.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
630
+ "model.decoder.layers.9.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
631
+ "model.decoder.layers.9.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
632
+ "model.encoder.layer_norm.bias": "model-00002-of-00003.safetensors",
633
+ "model.encoder.layer_norm.weight": "model-00002-of-00003.safetensors",
634
+ "model.encoder.layers.0.fc1.bias": "model-00001-of-00003.safetensors",
635
+ "model.encoder.layers.0.fc1.weight": "model-00001-of-00003.safetensors",
636
+ "model.encoder.layers.0.fc2.bias": "model-00001-of-00003.safetensors",
637
+ "model.encoder.layers.0.fc2.weight": "model-00001-of-00003.safetensors",
638
+ "model.encoder.layers.0.final_layer_norm.bias": "model-00001-of-00003.safetensors",
639
+ "model.encoder.layers.0.final_layer_norm.weight": "model-00001-of-00003.safetensors",
640
+ "model.encoder.layers.0.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
641
+ "model.encoder.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
642
+ "model.encoder.layers.0.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
643
+ "model.encoder.layers.0.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
644
+ "model.encoder.layers.0.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
645
+ "model.encoder.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
646
+ "model.encoder.layers.0.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
647
+ "model.encoder.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
648
+ "model.encoder.layers.0.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
649
+ "model.encoder.layers.0.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
650
+ "model.encoder.layers.1.fc1.bias": "model-00001-of-00003.safetensors",
651
+ "model.encoder.layers.1.fc1.weight": "model-00001-of-00003.safetensors",
652
+ "model.encoder.layers.1.fc2.bias": "model-00001-of-00003.safetensors",
653
+ "model.encoder.layers.1.fc2.weight": "model-00001-of-00003.safetensors",
654
+ "model.encoder.layers.1.final_layer_norm.bias": "model-00001-of-00003.safetensors",
655
+ "model.encoder.layers.1.final_layer_norm.weight": "model-00001-of-00003.safetensors",
656
+ "model.encoder.layers.1.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
657
+ "model.encoder.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
658
+ "model.encoder.layers.1.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
659
+ "model.encoder.layers.1.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
660
+ "model.encoder.layers.1.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
661
+ "model.encoder.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
662
+ "model.encoder.layers.1.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
663
+ "model.encoder.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
664
+ "model.encoder.layers.1.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
665
+ "model.encoder.layers.1.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
666
+ "model.encoder.layers.10.fc1.bias": "model-00001-of-00003.safetensors",
667
+ "model.encoder.layers.10.fc1.weight": "model-00001-of-00003.safetensors",
668
+ "model.encoder.layers.10.fc2.bias": "model-00001-of-00003.safetensors",
669
+ "model.encoder.layers.10.fc2.weight": "model-00001-of-00003.safetensors",
670
+ "model.encoder.layers.10.final_layer_norm.bias": "model-00001-of-00003.safetensors",
671
+ "model.encoder.layers.10.final_layer_norm.weight": "model-00001-of-00003.safetensors",
672
+ "model.encoder.layers.10.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
673
+ "model.encoder.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
674
+ "model.encoder.layers.10.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
675
+ "model.encoder.layers.10.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
676
+ "model.encoder.layers.10.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
677
+ "model.encoder.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
678
+ "model.encoder.layers.10.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
679
+ "model.encoder.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
680
+ "model.encoder.layers.10.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
681
+ "model.encoder.layers.10.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
682
+ "model.encoder.layers.11.fc1.bias": "model-00001-of-00003.safetensors",
683
+ "model.encoder.layers.11.fc1.weight": "model-00001-of-00003.safetensors",
684
+ "model.encoder.layers.11.fc2.bias": "model-00001-of-00003.safetensors",
685
+ "model.encoder.layers.11.fc2.weight": "model-00001-of-00003.safetensors",
686
+ "model.encoder.layers.11.final_layer_norm.bias": "model-00001-of-00003.safetensors",
687
+ "model.encoder.layers.11.final_layer_norm.weight": "model-00001-of-00003.safetensors",
688
+ "model.encoder.layers.11.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
689
+ "model.encoder.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
690
+ "model.encoder.layers.11.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
691
+ "model.encoder.layers.11.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
692
+ "model.encoder.layers.11.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
693
+ "model.encoder.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
694
+ "model.encoder.layers.11.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
695
+ "model.encoder.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
696
+ "model.encoder.layers.11.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
697
+ "model.encoder.layers.11.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
698
+ "model.encoder.layers.12.fc1.bias": "model-00001-of-00003.safetensors",
699
+ "model.encoder.layers.12.fc1.weight": "model-00001-of-00003.safetensors",
700
+ "model.encoder.layers.12.fc2.bias": "model-00001-of-00003.safetensors",
701
+ "model.encoder.layers.12.fc2.weight": "model-00001-of-00003.safetensors",
702
+ "model.encoder.layers.12.final_layer_norm.bias": "model-00001-of-00003.safetensors",
703
+ "model.encoder.layers.12.final_layer_norm.weight": "model-00001-of-00003.safetensors",
704
+ "model.encoder.layers.12.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
705
+ "model.encoder.layers.12.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
706
+ "model.encoder.layers.12.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
707
+ "model.encoder.layers.12.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
708
+ "model.encoder.layers.12.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
709
+ "model.encoder.layers.12.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
710
+ "model.encoder.layers.12.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
711
+ "model.encoder.layers.12.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
712
+ "model.encoder.layers.12.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
713
+ "model.encoder.layers.12.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
714
+ "model.encoder.layers.13.fc1.bias": "model-00001-of-00003.safetensors",
715
+ "model.encoder.layers.13.fc1.weight": "model-00001-of-00003.safetensors",
716
+ "model.encoder.layers.13.fc2.bias": "model-00001-of-00003.safetensors",
717
+ "model.encoder.layers.13.fc2.weight": "model-00001-of-00003.safetensors",
718
+ "model.encoder.layers.13.final_layer_norm.bias": "model-00001-of-00003.safetensors",
719
+ "model.encoder.layers.13.final_layer_norm.weight": "model-00001-of-00003.safetensors",
720
+ "model.encoder.layers.13.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
721
+ "model.encoder.layers.13.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
722
+ "model.encoder.layers.13.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
723
+ "model.encoder.layers.13.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
724
+ "model.encoder.layers.13.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
725
+ "model.encoder.layers.13.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
726
+ "model.encoder.layers.13.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
727
+ "model.encoder.layers.13.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
728
+ "model.encoder.layers.13.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
729
+ "model.encoder.layers.13.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
730
+ "model.encoder.layers.14.fc1.bias": "model-00002-of-00003.safetensors",
731
+ "model.encoder.layers.14.fc1.weight": "model-00002-of-00003.safetensors",
732
+ "model.encoder.layers.14.fc2.bias": "model-00002-of-00003.safetensors",
733
+ "model.encoder.layers.14.fc2.weight": "model-00002-of-00003.safetensors",
734
+ "model.encoder.layers.14.final_layer_norm.bias": "model-00002-of-00003.safetensors",
735
+ "model.encoder.layers.14.final_layer_norm.weight": "model-00002-of-00003.safetensors",
736
+ "model.encoder.layers.14.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
737
+ "model.encoder.layers.14.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
738
+ "model.encoder.layers.14.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
739
+ "model.encoder.layers.14.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
740
+ "model.encoder.layers.14.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
741
+ "model.encoder.layers.14.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
742
+ "model.encoder.layers.14.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
743
+ "model.encoder.layers.14.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
744
+ "model.encoder.layers.14.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
745
+ "model.encoder.layers.14.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
746
+ "model.encoder.layers.15.fc1.bias": "model-00002-of-00003.safetensors",
747
+ "model.encoder.layers.15.fc1.weight": "model-00002-of-00003.safetensors",
748
+ "model.encoder.layers.15.fc2.bias": "model-00002-of-00003.safetensors",
749
+ "model.encoder.layers.15.fc2.weight": "model-00002-of-00003.safetensors",
750
+ "model.encoder.layers.15.final_layer_norm.bias": "model-00002-of-00003.safetensors",
751
+ "model.encoder.layers.15.final_layer_norm.weight": "model-00002-of-00003.safetensors",
752
+ "model.encoder.layers.15.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
753
+ "model.encoder.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
754
+ "model.encoder.layers.15.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
755
+ "model.encoder.layers.15.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
756
+ "model.encoder.layers.15.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
757
+ "model.encoder.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
758
+ "model.encoder.layers.15.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
759
+ "model.encoder.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
760
+ "model.encoder.layers.15.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
761
+ "model.encoder.layers.15.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
762
+ "model.encoder.layers.16.fc1.bias": "model-00002-of-00003.safetensors",
763
+ "model.encoder.layers.16.fc1.weight": "model-00002-of-00003.safetensors",
764
+ "model.encoder.layers.16.fc2.bias": "model-00002-of-00003.safetensors",
765
+ "model.encoder.layers.16.fc2.weight": "model-00002-of-00003.safetensors",
766
+ "model.encoder.layers.16.final_layer_norm.bias": "model-00002-of-00003.safetensors",
767
+ "model.encoder.layers.16.final_layer_norm.weight": "model-00002-of-00003.safetensors",
768
+ "model.encoder.layers.16.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
769
+ "model.encoder.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
770
+ "model.encoder.layers.16.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
771
+ "model.encoder.layers.16.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
772
+ "model.encoder.layers.16.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
773
+ "model.encoder.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
774
+ "model.encoder.layers.16.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
775
+ "model.encoder.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
776
+ "model.encoder.layers.16.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
777
+ "model.encoder.layers.16.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
778
+ "model.encoder.layers.17.fc1.bias": "model-00002-of-00003.safetensors",
779
+ "model.encoder.layers.17.fc1.weight": "model-00002-of-00003.safetensors",
780
+ "model.encoder.layers.17.fc2.bias": "model-00002-of-00003.safetensors",
781
+ "model.encoder.layers.17.fc2.weight": "model-00002-of-00003.safetensors",
782
+ "model.encoder.layers.17.final_layer_norm.bias": "model-00002-of-00003.safetensors",
783
+ "model.encoder.layers.17.final_layer_norm.weight": "model-00002-of-00003.safetensors",
784
+ "model.encoder.layers.17.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
785
+ "model.encoder.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
786
+ "model.encoder.layers.17.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
787
+ "model.encoder.layers.17.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
788
+ "model.encoder.layers.17.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
789
+ "model.encoder.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
790
+ "model.encoder.layers.17.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
791
+ "model.encoder.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
792
+ "model.encoder.layers.17.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
793
+ "model.encoder.layers.17.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
794
+ "model.encoder.layers.18.fc1.bias": "model-00002-of-00003.safetensors",
795
+ "model.encoder.layers.18.fc1.weight": "model-00002-of-00003.safetensors",
796
+ "model.encoder.layers.18.fc2.bias": "model-00002-of-00003.safetensors",
797
+ "model.encoder.layers.18.fc2.weight": "model-00002-of-00003.safetensors",
798
+ "model.encoder.layers.18.final_layer_norm.bias": "model-00002-of-00003.safetensors",
799
+ "model.encoder.layers.18.final_layer_norm.weight": "model-00002-of-00003.safetensors",
800
+ "model.encoder.layers.18.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
801
+ "model.encoder.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
802
+ "model.encoder.layers.18.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
803
+ "model.encoder.layers.18.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
804
+ "model.encoder.layers.18.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
805
+ "model.encoder.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
806
+ "model.encoder.layers.18.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
807
+ "model.encoder.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
808
+ "model.encoder.layers.18.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
809
+ "model.encoder.layers.18.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
810
+ "model.encoder.layers.19.fc1.bias": "model-00002-of-00003.safetensors",
811
+ "model.encoder.layers.19.fc1.weight": "model-00002-of-00003.safetensors",
812
+ "model.encoder.layers.19.fc2.bias": "model-00002-of-00003.safetensors",
813
+ "model.encoder.layers.19.fc2.weight": "model-00002-of-00003.safetensors",
814
+ "model.encoder.layers.19.final_layer_norm.bias": "model-00002-of-00003.safetensors",
815
+ "model.encoder.layers.19.final_layer_norm.weight": "model-00002-of-00003.safetensors",
816
+ "model.encoder.layers.19.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
817
+ "model.encoder.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
818
+ "model.encoder.layers.19.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
819
+ "model.encoder.layers.19.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
820
+ "model.encoder.layers.19.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
821
+ "model.encoder.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
822
+ "model.encoder.layers.19.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
823
+ "model.encoder.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
824
+ "model.encoder.layers.19.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
825
+ "model.encoder.layers.19.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
826
+ "model.encoder.layers.2.fc1.bias": "model-00001-of-00003.safetensors",
827
+ "model.encoder.layers.2.fc1.weight": "model-00001-of-00003.safetensors",
828
+ "model.encoder.layers.2.fc2.bias": "model-00001-of-00003.safetensors",
829
+ "model.encoder.layers.2.fc2.weight": "model-00001-of-00003.safetensors",
830
+ "model.encoder.layers.2.final_layer_norm.bias": "model-00001-of-00003.safetensors",
831
+ "model.encoder.layers.2.final_layer_norm.weight": "model-00001-of-00003.safetensors",
832
+ "model.encoder.layers.2.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
833
+ "model.encoder.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
834
+ "model.encoder.layers.2.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
835
+ "model.encoder.layers.2.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
836
+ "model.encoder.layers.2.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
837
+ "model.encoder.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
838
+ "model.encoder.layers.2.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
839
+ "model.encoder.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
840
+ "model.encoder.layers.2.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
841
+ "model.encoder.layers.2.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
842
+ "model.encoder.layers.20.fc1.bias": "model-00002-of-00003.safetensors",
843
+ "model.encoder.layers.20.fc1.weight": "model-00002-of-00003.safetensors",
844
+ "model.encoder.layers.20.fc2.bias": "model-00002-of-00003.safetensors",
845
+ "model.encoder.layers.20.fc2.weight": "model-00002-of-00003.safetensors",
846
+ "model.encoder.layers.20.final_layer_norm.bias": "model-00002-of-00003.safetensors",
847
+ "model.encoder.layers.20.final_layer_norm.weight": "model-00002-of-00003.safetensors",
848
+ "model.encoder.layers.20.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
849
+ "model.encoder.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
850
+ "model.encoder.layers.20.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
851
+ "model.encoder.layers.20.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
852
+ "model.encoder.layers.20.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
853
+ "model.encoder.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
854
+ "model.encoder.layers.20.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
855
+ "model.encoder.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
856
+ "model.encoder.layers.20.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
857
+ "model.encoder.layers.20.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
858
+ "model.encoder.layers.21.fc1.bias": "model-00002-of-00003.safetensors",
859
+ "model.encoder.layers.21.fc1.weight": "model-00002-of-00003.safetensors",
860
+ "model.encoder.layers.21.fc2.bias": "model-00002-of-00003.safetensors",
861
+ "model.encoder.layers.21.fc2.weight": "model-00002-of-00003.safetensors",
862
+ "model.encoder.layers.21.final_layer_norm.bias": "model-00002-of-00003.safetensors",
863
+ "model.encoder.layers.21.final_layer_norm.weight": "model-00002-of-00003.safetensors",
864
+ "model.encoder.layers.21.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
865
+ "model.encoder.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
866
+ "model.encoder.layers.21.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
867
+ "model.encoder.layers.21.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
868
+ "model.encoder.layers.21.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
869
+ "model.encoder.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
870
+ "model.encoder.layers.21.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
871
+ "model.encoder.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
872
+ "model.encoder.layers.21.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
873
+ "model.encoder.layers.21.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
874
+ "model.encoder.layers.22.fc1.bias": "model-00002-of-00003.safetensors",
875
+ "model.encoder.layers.22.fc1.weight": "model-00002-of-00003.safetensors",
876
+ "model.encoder.layers.22.fc2.bias": "model-00002-of-00003.safetensors",
877
+ "model.encoder.layers.22.fc2.weight": "model-00002-of-00003.safetensors",
878
+ "model.encoder.layers.22.final_layer_norm.bias": "model-00002-of-00003.safetensors",
879
+ "model.encoder.layers.22.final_layer_norm.weight": "model-00002-of-00003.safetensors",
880
+ "model.encoder.layers.22.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
881
+ "model.encoder.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
882
+ "model.encoder.layers.22.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
883
+ "model.encoder.layers.22.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
884
+ "model.encoder.layers.22.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
885
+ "model.encoder.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
886
+ "model.encoder.layers.22.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
887
+ "model.encoder.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
888
+ "model.encoder.layers.22.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
889
+ "model.encoder.layers.22.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
890
+ "model.encoder.layers.23.fc1.bias": "model-00002-of-00003.safetensors",
891
+ "model.encoder.layers.23.fc1.weight": "model-00002-of-00003.safetensors",
892
+ "model.encoder.layers.23.fc2.bias": "model-00002-of-00003.safetensors",
893
+ "model.encoder.layers.23.fc2.weight": "model-00002-of-00003.safetensors",
894
+ "model.encoder.layers.23.final_layer_norm.bias": "model-00002-of-00003.safetensors",
895
+ "model.encoder.layers.23.final_layer_norm.weight": "model-00002-of-00003.safetensors",
896
+ "model.encoder.layers.23.self_attn.k_proj.bias": "model-00002-of-00003.safetensors",
897
+ "model.encoder.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
898
+ "model.encoder.layers.23.self_attn.out_proj.bias": "model-00002-of-00003.safetensors",
899
+ "model.encoder.layers.23.self_attn.out_proj.weight": "model-00002-of-00003.safetensors",
900
+ "model.encoder.layers.23.self_attn.q_proj.bias": "model-00002-of-00003.safetensors",
901
+ "model.encoder.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
902
+ "model.encoder.layers.23.self_attn.v_proj.bias": "model-00002-of-00003.safetensors",
903
+ "model.encoder.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
904
+ "model.encoder.layers.23.self_attn_layer_norm.bias": "model-00002-of-00003.safetensors",
905
+ "model.encoder.layers.23.self_attn_layer_norm.weight": "model-00002-of-00003.safetensors",
906
+ "model.encoder.layers.3.fc1.bias": "model-00001-of-00003.safetensors",
907
+ "model.encoder.layers.3.fc1.weight": "model-00001-of-00003.safetensors",
908
+ "model.encoder.layers.3.fc2.bias": "model-00001-of-00003.safetensors",
909
+ "model.encoder.layers.3.fc2.weight": "model-00001-of-00003.safetensors",
910
+ "model.encoder.layers.3.final_layer_norm.bias": "model-00001-of-00003.safetensors",
911
+ "model.encoder.layers.3.final_layer_norm.weight": "model-00001-of-00003.safetensors",
912
+ "model.encoder.layers.3.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
913
+ "model.encoder.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
914
+ "model.encoder.layers.3.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
915
+ "model.encoder.layers.3.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
916
+ "model.encoder.layers.3.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
917
+ "model.encoder.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
918
+ "model.encoder.layers.3.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
919
+ "model.encoder.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
920
+ "model.encoder.layers.3.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
921
+ "model.encoder.layers.3.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
922
+ "model.encoder.layers.4.fc1.bias": "model-00001-of-00003.safetensors",
923
+ "model.encoder.layers.4.fc1.weight": "model-00001-of-00003.safetensors",
924
+ "model.encoder.layers.4.fc2.bias": "model-00001-of-00003.safetensors",
925
+ "model.encoder.layers.4.fc2.weight": "model-00001-of-00003.safetensors",
926
+ "model.encoder.layers.4.final_layer_norm.bias": "model-00001-of-00003.safetensors",
927
+ "model.encoder.layers.4.final_layer_norm.weight": "model-00001-of-00003.safetensors",
928
+ "model.encoder.layers.4.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
929
+ "model.encoder.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
930
+ "model.encoder.layers.4.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
931
+ "model.encoder.layers.4.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
932
+ "model.encoder.layers.4.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
933
+ "model.encoder.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
934
+ "model.encoder.layers.4.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
935
+ "model.encoder.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
936
+ "model.encoder.layers.4.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
937
+ "model.encoder.layers.4.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
938
+ "model.encoder.layers.5.fc1.bias": "model-00001-of-00003.safetensors",
939
+ "model.encoder.layers.5.fc1.weight": "model-00001-of-00003.safetensors",
940
+ "model.encoder.layers.5.fc2.bias": "model-00001-of-00003.safetensors",
941
+ "model.encoder.layers.5.fc2.weight": "model-00001-of-00003.safetensors",
942
+ "model.encoder.layers.5.final_layer_norm.bias": "model-00001-of-00003.safetensors",
943
+ "model.encoder.layers.5.final_layer_norm.weight": "model-00001-of-00003.safetensors",
944
+ "model.encoder.layers.5.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
945
+ "model.encoder.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
946
+ "model.encoder.layers.5.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
947
+ "model.encoder.layers.5.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
948
+ "model.encoder.layers.5.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
949
+ "model.encoder.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
950
+ "model.encoder.layers.5.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
951
+ "model.encoder.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
952
+ "model.encoder.layers.5.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
953
+ "model.encoder.layers.5.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
954
+ "model.encoder.layers.6.fc1.bias": "model-00001-of-00003.safetensors",
955
+ "model.encoder.layers.6.fc1.weight": "model-00001-of-00003.safetensors",
956
+ "model.encoder.layers.6.fc2.bias": "model-00001-of-00003.safetensors",
957
+ "model.encoder.layers.6.fc2.weight": "model-00001-of-00003.safetensors",
958
+ "model.encoder.layers.6.final_layer_norm.bias": "model-00001-of-00003.safetensors",
959
+ "model.encoder.layers.6.final_layer_norm.weight": "model-00001-of-00003.safetensors",
960
+ "model.encoder.layers.6.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
961
+ "model.encoder.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
962
+ "model.encoder.layers.6.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
963
+ "model.encoder.layers.6.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
964
+ "model.encoder.layers.6.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
965
+ "model.encoder.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
966
+ "model.encoder.layers.6.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
967
+ "model.encoder.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
968
+ "model.encoder.layers.6.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
969
+ "model.encoder.layers.6.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
970
+ "model.encoder.layers.7.fc1.bias": "model-00001-of-00003.safetensors",
971
+ "model.encoder.layers.7.fc1.weight": "model-00001-of-00003.safetensors",
972
+ "model.encoder.layers.7.fc2.bias": "model-00001-of-00003.safetensors",
973
+ "model.encoder.layers.7.fc2.weight": "model-00001-of-00003.safetensors",
974
+ "model.encoder.layers.7.final_layer_norm.bias": "model-00001-of-00003.safetensors",
975
+ "model.encoder.layers.7.final_layer_norm.weight": "model-00001-of-00003.safetensors",
976
+ "model.encoder.layers.7.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
977
+ "model.encoder.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
978
+ "model.encoder.layers.7.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
979
+ "model.encoder.layers.7.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
980
+ "model.encoder.layers.7.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
981
+ "model.encoder.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
982
+ "model.encoder.layers.7.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
983
+ "model.encoder.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
984
+ "model.encoder.layers.7.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
985
+ "model.encoder.layers.7.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
986
+ "model.encoder.layers.8.fc1.bias": "model-00001-of-00003.safetensors",
987
+ "model.encoder.layers.8.fc1.weight": "model-00001-of-00003.safetensors",
988
+ "model.encoder.layers.8.fc2.bias": "model-00001-of-00003.safetensors",
989
+ "model.encoder.layers.8.fc2.weight": "model-00001-of-00003.safetensors",
990
+ "model.encoder.layers.8.final_layer_norm.bias": "model-00001-of-00003.safetensors",
991
+ "model.encoder.layers.8.final_layer_norm.weight": "model-00001-of-00003.safetensors",
992
+ "model.encoder.layers.8.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
993
+ "model.encoder.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
994
+ "model.encoder.layers.8.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
995
+ "model.encoder.layers.8.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
996
+ "model.encoder.layers.8.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
997
+ "model.encoder.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
998
+ "model.encoder.layers.8.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
999
+ "model.encoder.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
1000
+ "model.encoder.layers.8.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
1001
+ "model.encoder.layers.8.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
1002
+ "model.encoder.layers.9.fc1.bias": "model-00001-of-00003.safetensors",
1003
+ "model.encoder.layers.9.fc1.weight": "model-00001-of-00003.safetensors",
1004
+ "model.encoder.layers.9.fc2.bias": "model-00001-of-00003.safetensors",
1005
+ "model.encoder.layers.9.fc2.weight": "model-00001-of-00003.safetensors",
1006
+ "model.encoder.layers.9.final_layer_norm.bias": "model-00001-of-00003.safetensors",
1007
+ "model.encoder.layers.9.final_layer_norm.weight": "model-00001-of-00003.safetensors",
1008
+ "model.encoder.layers.9.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
1009
+ "model.encoder.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
1010
+ "model.encoder.layers.9.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
1011
+ "model.encoder.layers.9.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
1012
+ "model.encoder.layers.9.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
1013
+ "model.encoder.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
1014
+ "model.encoder.layers.9.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
1015
+ "model.encoder.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
1016
+ "model.encoder.layers.9.self_attn_layer_norm.bias": "model-00001-of-00003.safetensors",
1017
+ "model.encoder.layers.9.self_attn_layer_norm.weight": "model-00001-of-00003.safetensors",
1018
+ "model.shared.weight": "model-00001-of-00003.safetensors"
1019
+ }
1020
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:964dfaa7ac6434ad31b410b5bb15cb8dffa605235f35b4d0959e1a643811eac8
3
+ size 26759792081
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bfd5dd94f886d69e3ad4a23eb307d763584c5f9f8e19ef59c17ed67f7ac76a5
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2632f8ed1b4919ce96b0877365030c7fe32365208ab4e76b058167660417dd8
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,2193 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.4561562514161417,
5
+ "eval_steps": 500,
6
+ "global_step": 54200,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.022658268001993928,
13
+ "grad_norm": 0.8800877332687378,
14
+ "learning_rate": 4.962236219996677e-05,
15
+ "loss": 0.2417,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.022658268001993928,
20
+ "eval_bleu": 30.5488,
21
+ "eval_chrf++": 57.639,
22
+ "eval_gen_len": 28.2956,
23
+ "eval_loss": 1.1721652746200562,
24
+ "eval_runtime": 777.8277,
25
+ "eval_samples_per_second": 3.214,
26
+ "eval_spbleu": 41.6724,
27
+ "eval_steps_per_second": 0.643,
28
+ "eval_ter": 56.0615,
29
+ "step": 500
30
+ },
31
+ {
32
+ "epoch": 0.045316536003987856,
33
+ "grad_norm": 0.8184657096862793,
34
+ "learning_rate": 4.924472439993354e-05,
35
+ "loss": 1.0084,
36
+ "step": 1000
37
+ },
38
+ {
39
+ "epoch": 0.045316536003987856,
40
+ "eval_bleu": 31.6027,
41
+ "eval_chrf++": 58.8947,
42
+ "eval_gen_len": 27.9,
43
+ "eval_loss": 1.0108660459518433,
44
+ "eval_runtime": 745.4422,
45
+ "eval_samples_per_second": 3.354,
46
+ "eval_spbleu": 44.064,
47
+ "eval_steps_per_second": 0.671,
48
+ "eval_ter": 52.7499,
49
+ "step": 1000
50
+ },
51
+ {
52
+ "epoch": 0.06797480400598178,
53
+ "grad_norm": 0.9998241662979126,
54
+ "learning_rate": 4.886708659990031e-05,
55
+ "loss": 1.0649,
56
+ "step": 1500
57
+ },
58
+ {
59
+ "epoch": 0.06797480400598178,
60
+ "eval_bleu": 35.962,
61
+ "eval_chrf++": 60.0663,
62
+ "eval_gen_len": 27.614,
63
+ "eval_loss": 0.9778443574905396,
64
+ "eval_runtime": 743.9627,
65
+ "eval_samples_per_second": 3.36,
66
+ "eval_spbleu": 45.4876,
67
+ "eval_steps_per_second": 0.672,
68
+ "eval_ter": 52.1752,
69
+ "step": 1500
70
+ },
71
+ {
72
+ "epoch": 0.09063307200797571,
73
+ "grad_norm": 0.9190363883972168,
74
+ "learning_rate": 4.848944879986708e-05,
75
+ "loss": 1.0414,
76
+ "step": 2000
77
+ },
78
+ {
79
+ "epoch": 0.09063307200797571,
80
+ "eval_bleu": 33.9141,
81
+ "eval_chrf++": 60.0862,
82
+ "eval_gen_len": 27.7216,
83
+ "eval_loss": 0.9621853232383728,
84
+ "eval_runtime": 739.7943,
85
+ "eval_samples_per_second": 3.379,
86
+ "eval_spbleu": 45.9477,
87
+ "eval_steps_per_second": 0.676,
88
+ "eval_ter": 51.1627,
89
+ "step": 2000
90
+ },
91
+ {
92
+ "epoch": 0.11329134000996964,
93
+ "grad_norm": 0.8132910132408142,
94
+ "learning_rate": 4.811181099983384e-05,
95
+ "loss": 1.0051,
96
+ "step": 2500
97
+ },
98
+ {
99
+ "epoch": 0.11329134000996964,
100
+ "eval_bleu": 33.3827,
101
+ "eval_chrf++": 60.7457,
102
+ "eval_gen_len": 28.04,
103
+ "eval_loss": 0.9484396576881409,
104
+ "eval_runtime": 751.6447,
105
+ "eval_samples_per_second": 3.326,
106
+ "eval_spbleu": 46.5548,
107
+ "eval_steps_per_second": 0.665,
108
+ "eval_ter": 50.9017,
109
+ "step": 2500
110
+ },
111
+ {
112
+ "epoch": 0.13594960801196357,
113
+ "grad_norm": 0.8168209195137024,
114
+ "learning_rate": 4.7734173199800606e-05,
115
+ "loss": 1.0033,
116
+ "step": 3000
117
+ },
118
+ {
119
+ "epoch": 0.13594960801196357,
120
+ "eval_bleu": 33.8725,
121
+ "eval_chrf++": 60.8653,
122
+ "eval_gen_len": 28.0696,
123
+ "eval_loss": 0.9424599409103394,
124
+ "eval_runtime": 759.5734,
125
+ "eval_samples_per_second": 3.291,
126
+ "eval_spbleu": 46.9775,
127
+ "eval_steps_per_second": 0.658,
128
+ "eval_ter": 50.6539,
129
+ "step": 3000
130
+ },
131
+ {
132
+ "epoch": 0.15860787601395748,
133
+ "grad_norm": 0.9191615581512451,
134
+ "learning_rate": 4.735653539976738e-05,
135
+ "loss": 0.994,
136
+ "step": 3500
137
+ },
138
+ {
139
+ "epoch": 0.15860787601395748,
140
+ "eval_bleu": 34.0874,
141
+ "eval_chrf++": 61.217,
142
+ "eval_gen_len": 27.996,
143
+ "eval_loss": 0.9314232468605042,
144
+ "eval_runtime": 754.8484,
145
+ "eval_samples_per_second": 3.312,
146
+ "eval_spbleu": 47.405,
147
+ "eval_steps_per_second": 0.662,
148
+ "eval_ter": 50.0527,
149
+ "step": 3500
150
+ },
151
+ {
152
+ "epoch": 0.18126614401595142,
153
+ "grad_norm": 0.7960318326950073,
154
+ "learning_rate": 4.697889759973415e-05,
155
+ "loss": 0.9801,
156
+ "step": 4000
157
+ },
158
+ {
159
+ "epoch": 0.18126614401595142,
160
+ "eval_bleu": 37.2884,
161
+ "eval_chrf++": 61.5163,
162
+ "eval_gen_len": 27.9868,
163
+ "eval_loss": 0.9198995232582092,
164
+ "eval_runtime": 750.5441,
165
+ "eval_samples_per_second": 3.331,
166
+ "eval_spbleu": 47.7096,
167
+ "eval_steps_per_second": 0.666,
168
+ "eval_ter": 50.0316,
169
+ "step": 4000
170
+ },
171
+ {
172
+ "epoch": 0.20392441201794534,
173
+ "grad_norm": 0.9299506545066833,
174
+ "learning_rate": 4.6601259799700914e-05,
175
+ "loss": 0.9679,
176
+ "step": 4500
177
+ },
178
+ {
179
+ "epoch": 0.20392441201794534,
180
+ "eval_bleu": 36.221,
181
+ "eval_chrf++": 61.6489,
182
+ "eval_gen_len": 27.7652,
183
+ "eval_loss": 0.9132654070854187,
184
+ "eval_runtime": 742.4239,
185
+ "eval_samples_per_second": 3.367,
186
+ "eval_spbleu": 48.0051,
187
+ "eval_steps_per_second": 0.673,
188
+ "eval_ter": 49.4911,
189
+ "step": 4500
190
+ },
191
+ {
192
+ "epoch": 0.22658268001993928,
193
+ "grad_norm": 0.8396582007408142,
194
+ "learning_rate": 4.622362199966768e-05,
195
+ "loss": 0.9567,
196
+ "step": 5000
197
+ },
198
+ {
199
+ "epoch": 0.22658268001993928,
200
+ "eval_bleu": 35.8613,
201
+ "eval_chrf++": 62.1554,
202
+ "eval_gen_len": 27.9184,
203
+ "eval_loss": 0.9109494090080261,
204
+ "eval_runtime": 749.1424,
205
+ "eval_samples_per_second": 3.337,
206
+ "eval_spbleu": 48.6507,
207
+ "eval_steps_per_second": 0.667,
208
+ "eval_ter": 49.9394,
209
+ "step": 5000
210
+ },
211
+ {
212
+ "epoch": 0.2492409480219332,
213
+ "grad_norm": 0.6273393034934998,
214
+ "learning_rate": 4.584598419963445e-05,
215
+ "loss": 0.9625,
216
+ "step": 5500
217
+ },
218
+ {
219
+ "epoch": 0.2492409480219332,
220
+ "eval_bleu": 34.9284,
221
+ "eval_chrf++": 61.8771,
222
+ "eval_gen_len": 27.8456,
223
+ "eval_loss": 0.9041927456855774,
224
+ "eval_runtime": 762.3368,
225
+ "eval_samples_per_second": 3.279,
226
+ "eval_spbleu": 48.461,
227
+ "eval_steps_per_second": 0.656,
228
+ "eval_ter": 49.4832,
229
+ "step": 5500
230
+ },
231
+ {
232
+ "epoch": 0.27189921602392714,
233
+ "grad_norm": 0.8030633330345154,
234
+ "learning_rate": 4.546834639960122e-05,
235
+ "loss": 0.9465,
236
+ "step": 6000
237
+ },
238
+ {
239
+ "epoch": 0.27189921602392714,
240
+ "eval_bleu": 35.4623,
241
+ "eval_chrf++": 62.3793,
242
+ "eval_gen_len": 27.7244,
243
+ "eval_loss": 0.8957546949386597,
244
+ "eval_runtime": 760.5666,
245
+ "eval_samples_per_second": 3.287,
246
+ "eval_spbleu": 49.0727,
247
+ "eval_steps_per_second": 0.657,
248
+ "eval_ter": 48.5367,
249
+ "step": 6000
250
+ },
251
+ {
252
+ "epoch": 0.2945574840259211,
253
+ "grad_norm": 0.8178768754005432,
254
+ "learning_rate": 4.509070859956798e-05,
255
+ "loss": 0.9275,
256
+ "step": 6500
257
+ },
258
+ {
259
+ "epoch": 0.2945574840259211,
260
+ "eval_bleu": 35.2431,
261
+ "eval_chrf++": 62.0683,
262
+ "eval_gen_len": 27.8264,
263
+ "eval_loss": 0.8866144418716431,
264
+ "eval_runtime": 751.0343,
265
+ "eval_samples_per_second": 3.329,
266
+ "eval_spbleu": 48.8382,
267
+ "eval_steps_per_second": 0.666,
268
+ "eval_ter": 48.9164,
269
+ "step": 6500
270
+ },
271
+ {
272
+ "epoch": 0.31721575202791497,
273
+ "grad_norm": 0.6623912453651428,
274
+ "learning_rate": 4.471307079953475e-05,
275
+ "loss": 0.925,
276
+ "step": 7000
277
+ },
278
+ {
279
+ "epoch": 0.31721575202791497,
280
+ "eval_bleu": 35.6474,
281
+ "eval_chrf++": 62.5261,
282
+ "eval_gen_len": 27.9388,
283
+ "eval_loss": 0.884535551071167,
284
+ "eval_runtime": 758.4604,
285
+ "eval_samples_per_second": 3.296,
286
+ "eval_spbleu": 49.2377,
287
+ "eval_steps_per_second": 0.659,
288
+ "eval_ter": 48.4945,
289
+ "step": 7000
290
+ },
291
+ {
292
+ "epoch": 0.3398740200299089,
293
+ "grad_norm": 0.8043058514595032,
294
+ "learning_rate": 4.433543299950152e-05,
295
+ "loss": 0.928,
296
+ "step": 7500
297
+ },
298
+ {
299
+ "epoch": 0.3398740200299089,
300
+ "eval_bleu": 35.1147,
301
+ "eval_chrf++": 62.2742,
302
+ "eval_gen_len": 28.0044,
303
+ "eval_loss": 0.8792969584465027,
304
+ "eval_runtime": 755.7098,
305
+ "eval_samples_per_second": 3.308,
306
+ "eval_spbleu": 48.8554,
307
+ "eval_steps_per_second": 0.662,
308
+ "eval_ter": 49.3198,
309
+ "step": 7500
310
+ },
311
+ {
312
+ "epoch": 0.36253228803190285,
313
+ "grad_norm": 0.747755765914917,
314
+ "learning_rate": 4.395779519946829e-05,
315
+ "loss": 0.9096,
316
+ "step": 8000
317
+ },
318
+ {
319
+ "epoch": 0.36253228803190285,
320
+ "eval_bleu": 35.7901,
321
+ "eval_chrf++": 62.8302,
322
+ "eval_gen_len": 27.9312,
323
+ "eval_loss": 0.8780434727668762,
324
+ "eval_runtime": 770.928,
325
+ "eval_samples_per_second": 3.243,
326
+ "eval_spbleu": 49.5562,
327
+ "eval_steps_per_second": 0.649,
328
+ "eval_ter": 48.5578,
329
+ "step": 8000
330
+ },
331
+ {
332
+ "epoch": 0.3851905560338968,
333
+ "grad_norm": 0.6910504102706909,
334
+ "learning_rate": 4.358015739943506e-05,
335
+ "loss": 0.9014,
336
+ "step": 8500
337
+ },
338
+ {
339
+ "epoch": 0.3851905560338968,
340
+ "eval_bleu": 36.2166,
341
+ "eval_chrf++": 63.0932,
342
+ "eval_gen_len": 27.8432,
343
+ "eval_loss": 0.8704683780670166,
344
+ "eval_runtime": 750.7247,
345
+ "eval_samples_per_second": 3.33,
346
+ "eval_spbleu": 50.2483,
347
+ "eval_steps_per_second": 0.666,
348
+ "eval_ter": 48.3258,
349
+ "step": 8500
350
+ },
351
+ {
352
+ "epoch": 0.4078488240358907,
353
+ "grad_norm": 0.7716640830039978,
354
+ "learning_rate": 4.3202519599401825e-05,
355
+ "loss": 0.9059,
356
+ "step": 9000
357
+ },
358
+ {
359
+ "epoch": 0.4078488240358907,
360
+ "eval_bleu": 41.2195,
361
+ "eval_chrf++": 62.902,
362
+ "eval_gen_len": 27.5208,
363
+ "eval_loss": 0.8670679926872253,
364
+ "eval_runtime": 748.217,
365
+ "eval_samples_per_second": 3.341,
366
+ "eval_spbleu": 49.6619,
367
+ "eval_steps_per_second": 0.668,
368
+ "eval_ter": 48.3732,
369
+ "step": 9000
370
+ },
371
+ {
372
+ "epoch": 0.4305070920378846,
373
+ "grad_norm": 0.801590085029602,
374
+ "learning_rate": 4.282488179936859e-05,
375
+ "loss": 0.8976,
376
+ "step": 9500
377
+ },
378
+ {
379
+ "epoch": 0.4305070920378846,
380
+ "eval_bleu": 35.9681,
381
+ "eval_chrf++": 63.0627,
382
+ "eval_gen_len": 28.004,
383
+ "eval_loss": 0.8662621378898621,
384
+ "eval_runtime": 762.3803,
385
+ "eval_samples_per_second": 3.279,
386
+ "eval_spbleu": 50.006,
387
+ "eval_steps_per_second": 0.656,
388
+ "eval_ter": 48.4945,
389
+ "step": 9500
390
+ },
391
+ {
392
+ "epoch": 0.45316536003987856,
393
+ "grad_norm": 0.7821282148361206,
394
+ "learning_rate": 4.244724399933536e-05,
395
+ "loss": 0.889,
396
+ "step": 10000
397
+ },
398
+ {
399
+ "epoch": 0.45316536003987856,
400
+ "eval_bleu": 36.3811,
401
+ "eval_chrf++": 63.2137,
402
+ "eval_gen_len": 27.9044,
403
+ "eval_loss": 0.8606961965560913,
404
+ "eval_runtime": 766.3054,
405
+ "eval_samples_per_second": 3.262,
406
+ "eval_spbleu": 50.2147,
407
+ "eval_steps_per_second": 0.652,
408
+ "eval_ter": 47.6614,
409
+ "step": 10000
410
+ },
411
+ {
412
+ "epoch": 0.4758236280418725,
413
+ "grad_norm": 0.7221835851669312,
414
+ "learning_rate": 4.206960619930213e-05,
415
+ "loss": 0.8979,
416
+ "step": 10500
417
+ },
418
+ {
419
+ "epoch": 0.4758236280418725,
420
+ "eval_bleu": 36.4697,
421
+ "eval_chrf++": 63.3942,
422
+ "eval_gen_len": 27.9276,
423
+ "eval_loss": 0.8558794856071472,
424
+ "eval_runtime": 758.2694,
425
+ "eval_samples_per_second": 3.297,
426
+ "eval_spbleu": 50.6637,
427
+ "eval_steps_per_second": 0.659,
428
+ "eval_ter": 48.0226,
429
+ "step": 10500
430
+ },
431
+ {
432
+ "epoch": 0.4984818960438664,
433
+ "grad_norm": 0.5793339014053345,
434
+ "learning_rate": 4.1691968399268894e-05,
435
+ "loss": 0.8817,
436
+ "step": 11000
437
+ },
438
+ {
439
+ "epoch": 0.4984818960438664,
440
+ "eval_bleu": 36.4549,
441
+ "eval_chrf++": 63.2444,
442
+ "eval_gen_len": 27.7924,
443
+ "eval_loss": 0.8521751761436462,
444
+ "eval_runtime": 746.8271,
445
+ "eval_samples_per_second": 3.347,
446
+ "eval_spbleu": 50.4999,
447
+ "eval_steps_per_second": 0.669,
448
+ "eval_ter": 47.9487,
449
+ "step": 11000
450
+ },
451
+ {
452
+ "epoch": 0.5211401640458604,
453
+ "grad_norm": 0.8384801149368286,
454
+ "learning_rate": 4.131433059923566e-05,
455
+ "loss": 0.8696,
456
+ "step": 11500
457
+ },
458
+ {
459
+ "epoch": 0.5211401640458604,
460
+ "eval_bleu": 36.1199,
461
+ "eval_chrf++": 63.343,
462
+ "eval_gen_len": 28.076,
463
+ "eval_loss": 0.8511990308761597,
464
+ "eval_runtime": 758.0827,
465
+ "eval_samples_per_second": 3.298,
466
+ "eval_spbleu": 50.1722,
467
+ "eval_steps_per_second": 0.66,
468
+ "eval_ter": 48.6316,
469
+ "step": 11500
470
+ },
471
+ {
472
+ "epoch": 0.5437984320478543,
473
+ "grad_norm": 0.6142855882644653,
474
+ "learning_rate": 4.093669279920243e-05,
475
+ "loss": 0.8758,
476
+ "step": 12000
477
+ },
478
+ {
479
+ "epoch": 0.5437984320478543,
480
+ "eval_bleu": 36.7344,
481
+ "eval_chrf++": 63.7589,
482
+ "eval_gen_len": 27.9936,
483
+ "eval_loss": 0.8446237444877625,
484
+ "eval_runtime": 749.403,
485
+ "eval_samples_per_second": 3.336,
486
+ "eval_spbleu": 50.8742,
487
+ "eval_steps_per_second": 0.667,
488
+ "eval_ter": 47.1841,
489
+ "step": 12000
490
+ },
491
+ {
492
+ "epoch": 0.5664567000498482,
493
+ "grad_norm": 0.5856760144233704,
494
+ "learning_rate": 4.05590549991692e-05,
495
+ "loss": 0.8793,
496
+ "step": 12500
497
+ },
498
+ {
499
+ "epoch": 0.5664567000498482,
500
+ "eval_bleu": 41.021,
501
+ "eval_chrf++": 63.6181,
502
+ "eval_gen_len": 27.5668,
503
+ "eval_loss": 0.8422514796257019,
504
+ "eval_runtime": 733.2302,
505
+ "eval_samples_per_second": 3.41,
506
+ "eval_spbleu": 50.6419,
507
+ "eval_steps_per_second": 0.682,
508
+ "eval_ter": 47.1499,
509
+ "step": 12500
510
+ },
511
+ {
512
+ "epoch": 0.5891149680518422,
513
+ "grad_norm": 0.6606504321098328,
514
+ "learning_rate": 4.018141719913597e-05,
515
+ "loss": 0.8794,
516
+ "step": 13000
517
+ },
518
+ {
519
+ "epoch": 0.5891149680518422,
520
+ "eval_bleu": 37.286,
521
+ "eval_chrf++": 63.8579,
522
+ "eval_gen_len": 27.8624,
523
+ "eval_loss": 0.8412001729011536,
524
+ "eval_runtime": 745.0766,
525
+ "eval_samples_per_second": 3.355,
526
+ "eval_spbleu": 51.2084,
527
+ "eval_steps_per_second": 0.671,
528
+ "eval_ter": 46.4538,
529
+ "step": 13000
530
+ },
531
+ {
532
+ "epoch": 0.611773236053836,
533
+ "grad_norm": 0.7332282066345215,
534
+ "learning_rate": 3.980377939910274e-05,
535
+ "loss": 0.8543,
536
+ "step": 13500
537
+ },
538
+ {
539
+ "epoch": 0.611773236053836,
540
+ "eval_bleu": 36.8567,
541
+ "eval_chrf++": 63.6306,
542
+ "eval_gen_len": 27.7904,
543
+ "eval_loss": 0.8333261013031006,
544
+ "eval_runtime": 744.0262,
545
+ "eval_samples_per_second": 3.36,
546
+ "eval_spbleu": 50.873,
547
+ "eval_steps_per_second": 0.672,
548
+ "eval_ter": 47.0998,
549
+ "step": 13500
550
+ },
551
+ {
552
+ "epoch": 0.6344315040558299,
553
+ "grad_norm": 0.8010419607162476,
554
+ "learning_rate": 3.9426141599069504e-05,
555
+ "loss": 0.8661,
556
+ "step": 14000
557
+ },
558
+ {
559
+ "epoch": 0.6344315040558299,
560
+ "eval_bleu": 36.9197,
561
+ "eval_chrf++": 63.5882,
562
+ "eval_gen_len": 27.8996,
563
+ "eval_loss": 0.8315057754516602,
564
+ "eval_runtime": 748.0384,
565
+ "eval_samples_per_second": 3.342,
566
+ "eval_spbleu": 50.798,
567
+ "eval_steps_per_second": 0.668,
568
+ "eval_ter": 46.8968,
569
+ "step": 14000
570
+ },
571
+ {
572
+ "epoch": 0.6570897720578239,
573
+ "grad_norm": 0.5245931148529053,
574
+ "learning_rate": 3.9048503799036265e-05,
575
+ "loss": 0.8556,
576
+ "step": 14500
577
+ },
578
+ {
579
+ "epoch": 0.6570897720578239,
580
+ "eval_bleu": 37.9259,
581
+ "eval_chrf++": 63.8176,
582
+ "eval_gen_len": 27.84,
583
+ "eval_loss": 0.8274693489074707,
584
+ "eval_runtime": 748.0441,
585
+ "eval_samples_per_second": 3.342,
586
+ "eval_spbleu": 51.0265,
587
+ "eval_steps_per_second": 0.668,
588
+ "eval_ter": 46.9205,
589
+ "step": 14500
590
+ },
591
+ {
592
+ "epoch": 0.6797480400598178,
593
+ "grad_norm": 0.7052202820777893,
594
+ "learning_rate": 3.867086599900304e-05,
595
+ "loss": 0.8635,
596
+ "step": 15000
597
+ },
598
+ {
599
+ "epoch": 0.6797480400598178,
600
+ "eval_bleu": 38.0929,
601
+ "eval_chrf++": 63.8837,
602
+ "eval_gen_len": 27.8216,
603
+ "eval_loss": 0.8276916146278381,
604
+ "eval_runtime": 744.6549,
605
+ "eval_samples_per_second": 3.357,
606
+ "eval_spbleu": 51.13,
607
+ "eval_steps_per_second": 0.671,
608
+ "eval_ter": 46.9099,
609
+ "step": 15000
610
+ },
611
+ {
612
+ "epoch": 0.7024063080618117,
613
+ "grad_norm": 0.7750408053398132,
614
+ "learning_rate": 3.8293228198969806e-05,
615
+ "loss": 0.8412,
616
+ "step": 15500
617
+ },
618
+ {
619
+ "epoch": 0.7024063080618117,
620
+ "eval_bleu": 38.2488,
621
+ "eval_chrf++": 63.9152,
622
+ "eval_gen_len": 27.7136,
623
+ "eval_loss": 0.8237889409065247,
624
+ "eval_runtime": 739.0003,
625
+ "eval_samples_per_second": 3.383,
626
+ "eval_spbleu": 51.3155,
627
+ "eval_steps_per_second": 0.677,
628
+ "eval_ter": 46.5224,
629
+ "step": 15500
630
+ },
631
+ {
632
+ "epoch": 0.7250645760638057,
633
+ "grad_norm": 0.7442999482154846,
634
+ "learning_rate": 3.791559039893657e-05,
635
+ "loss": 0.8476,
636
+ "step": 16000
637
+ },
638
+ {
639
+ "epoch": 0.7250645760638057,
640
+ "eval_bleu": 40.1641,
641
+ "eval_chrf++": 64.262,
642
+ "eval_gen_len": 27.616,
643
+ "eval_loss": 0.8218015432357788,
644
+ "eval_runtime": 736.1573,
645
+ "eval_samples_per_second": 3.396,
646
+ "eval_spbleu": 51.5374,
647
+ "eval_steps_per_second": 0.679,
648
+ "eval_ter": 46.2982,
649
+ "step": 16000
650
+ },
651
+ {
652
+ "epoch": 0.7477228440657996,
653
+ "grad_norm": 0.6437325477600098,
654
+ "learning_rate": 3.753795259890334e-05,
655
+ "loss": 0.854,
656
+ "step": 16500
657
+ },
658
+ {
659
+ "epoch": 0.7477228440657996,
660
+ "eval_bleu": 37.6219,
661
+ "eval_chrf++": 64.4048,
662
+ "eval_gen_len": 27.9068,
663
+ "eval_loss": 0.8224019408226013,
664
+ "eval_runtime": 747.0111,
665
+ "eval_samples_per_second": 3.347,
666
+ "eval_spbleu": 51.8679,
667
+ "eval_steps_per_second": 0.669,
668
+ "eval_ter": 46.3009,
669
+ "step": 16500
670
+ },
671
+ {
672
+ "epoch": 0.7703811120677936,
673
+ "grad_norm": 0.7045587301254272,
674
+ "learning_rate": 3.716031479887011e-05,
675
+ "loss": 0.8433,
676
+ "step": 17000
677
+ },
678
+ {
679
+ "epoch": 0.7703811120677936,
680
+ "eval_bleu": 38.7383,
681
+ "eval_chrf++": 64.511,
682
+ "eval_gen_len": 27.94,
683
+ "eval_loss": 0.8197815418243408,
684
+ "eval_runtime": 750.586,
685
+ "eval_samples_per_second": 3.331,
686
+ "eval_spbleu": 51.9857,
687
+ "eval_steps_per_second": 0.666,
688
+ "eval_ter": 46.3378,
689
+ "step": 17000
690
+ },
691
+ {
692
+ "epoch": 0.7930393800697875,
693
+ "grad_norm": 0.6813265681266785,
694
+ "learning_rate": 3.678267699883688e-05,
695
+ "loss": 0.8372,
696
+ "step": 17500
697
+ },
698
+ {
699
+ "epoch": 0.7930393800697875,
700
+ "eval_bleu": 39.4841,
701
+ "eval_chrf++": 64.344,
702
+ "eval_gen_len": 27.5488,
703
+ "eval_loss": 0.8175507187843323,
704
+ "eval_runtime": 732.9073,
705
+ "eval_samples_per_second": 3.411,
706
+ "eval_spbleu": 51.5575,
707
+ "eval_steps_per_second": 0.682,
708
+ "eval_ter": 46.1163,
709
+ "step": 17500
710
+ },
711
+ {
712
+ "epoch": 0.8156976480717814,
713
+ "grad_norm": 0.5977945327758789,
714
+ "learning_rate": 3.640503919880365e-05,
715
+ "loss": 0.8317,
716
+ "step": 18000
717
+ },
718
+ {
719
+ "epoch": 0.8156976480717814,
720
+ "eval_bleu": 37.7506,
721
+ "eval_chrf++": 64.0896,
722
+ "eval_gen_len": 27.8936,
723
+ "eval_loss": 0.8134418725967407,
724
+ "eval_runtime": 746.7872,
725
+ "eval_samples_per_second": 3.348,
726
+ "eval_spbleu": 51.6484,
727
+ "eval_steps_per_second": 0.67,
728
+ "eval_ter": 46.6937,
729
+ "step": 18000
730
+ },
731
+ {
732
+ "epoch": 0.8383559160737754,
733
+ "grad_norm": 0.6788719892501831,
734
+ "learning_rate": 3.602740139877041e-05,
735
+ "loss": 0.8331,
736
+ "step": 18500
737
+ },
738
+ {
739
+ "epoch": 0.8383559160737754,
740
+ "eval_bleu": 37.9443,
741
+ "eval_chrf++": 64.6177,
742
+ "eval_gen_len": 28.022,
743
+ "eval_loss": 0.8157890439033508,
744
+ "eval_runtime": 748.194,
745
+ "eval_samples_per_second": 3.341,
746
+ "eval_spbleu": 52.1179,
747
+ "eval_steps_per_second": 0.668,
748
+ "eval_ter": 46.2323,
749
+ "step": 18500
750
+ },
751
+ {
752
+ "epoch": 0.8610141840757692,
753
+ "grad_norm": 0.6696301102638245,
754
+ "learning_rate": 3.5649763598737176e-05,
755
+ "loss": 0.8342,
756
+ "step": 19000
757
+ },
758
+ {
759
+ "epoch": 0.8610141840757692,
760
+ "eval_bleu": 38.7784,
761
+ "eval_chrf++": 64.6136,
762
+ "eval_gen_len": 27.9508,
763
+ "eval_loss": 0.8111441731452942,
764
+ "eval_runtime": 748.3754,
765
+ "eval_samples_per_second": 3.341,
766
+ "eval_spbleu": 52.0913,
767
+ "eval_steps_per_second": 0.668,
768
+ "eval_ter": 46.206,
769
+ "step": 19000
770
+ },
771
+ {
772
+ "epoch": 0.8836724520777631,
773
+ "grad_norm": 0.7288480401039124,
774
+ "learning_rate": 3.527212579870395e-05,
775
+ "loss": 0.8282,
776
+ "step": 19500
777
+ },
778
+ {
779
+ "epoch": 0.8836724520777631,
780
+ "eval_bleu": 38.1426,
781
+ "eval_chrf++": 64.6892,
782
+ "eval_gen_len": 27.8628,
783
+ "eval_loss": 0.8068262934684753,
784
+ "eval_runtime": 743.9866,
785
+ "eval_samples_per_second": 3.36,
786
+ "eval_spbleu": 52.3226,
787
+ "eval_steps_per_second": 0.672,
788
+ "eval_ter": 45.7841,
789
+ "step": 19500
790
+ },
791
+ {
792
+ "epoch": 0.9063307200797571,
793
+ "grad_norm": 0.6522256731987,
794
+ "learning_rate": 3.489448799867072e-05,
795
+ "loss": 0.8345,
796
+ "step": 20000
797
+ },
798
+ {
799
+ "epoch": 0.9063307200797571,
800
+ "eval_bleu": 37.3235,
801
+ "eval_chrf++": 64.2609,
802
+ "eval_gen_len": 27.9892,
803
+ "eval_loss": 0.8071653246879578,
804
+ "eval_runtime": 750.865,
805
+ "eval_samples_per_second": 3.329,
806
+ "eval_spbleu": 51.689,
807
+ "eval_steps_per_second": 0.666,
808
+ "eval_ter": 47.0022,
809
+ "step": 20000
810
+ },
811
+ {
812
+ "epoch": 0.928988988081751,
813
+ "grad_norm": 0.7145525217056274,
814
+ "learning_rate": 3.4516850198637484e-05,
815
+ "loss": 0.8224,
816
+ "step": 20500
817
+ },
818
+ {
819
+ "epoch": 0.928988988081751,
820
+ "eval_bleu": 37.5387,
821
+ "eval_chrf++": 64.2559,
822
+ "eval_gen_len": 27.82,
823
+ "eval_loss": 0.8012556433677673,
824
+ "eval_runtime": 768.5173,
825
+ "eval_samples_per_second": 3.253,
826
+ "eval_spbleu": 51.7469,
827
+ "eval_steps_per_second": 0.651,
828
+ "eval_ter": 46.3668,
829
+ "step": 20500
830
+ },
831
+ {
832
+ "epoch": 0.951647256083745,
833
+ "grad_norm": 0.7082468271255493,
834
+ "learning_rate": 3.413921239860425e-05,
835
+ "loss": 0.8315,
836
+ "step": 21000
837
+ },
838
+ {
839
+ "epoch": 0.951647256083745,
840
+ "eval_bleu": 37.9748,
841
+ "eval_chrf++": 64.7286,
842
+ "eval_gen_len": 27.8264,
843
+ "eval_loss": 0.7990391254425049,
844
+ "eval_runtime": 772.2382,
845
+ "eval_samples_per_second": 3.237,
846
+ "eval_spbleu": 52.3829,
847
+ "eval_steps_per_second": 0.647,
848
+ "eval_ter": 45.8711,
849
+ "step": 21000
850
+ },
851
+ {
852
+ "epoch": 0.9743055240857389,
853
+ "grad_norm": 0.6599904298782349,
854
+ "learning_rate": 3.376157459857102e-05,
855
+ "loss": 0.8269,
856
+ "step": 21500
857
+ },
858
+ {
859
+ "epoch": 0.9743055240857389,
860
+ "eval_bleu": 39.3032,
861
+ "eval_chrf++": 64.9609,
862
+ "eval_gen_len": 27.8436,
863
+ "eval_loss": 0.7974932789802551,
864
+ "eval_runtime": 771.5992,
865
+ "eval_samples_per_second": 3.24,
866
+ "eval_spbleu": 52.5588,
867
+ "eval_steps_per_second": 0.648,
868
+ "eval_ter": 45.8975,
869
+ "step": 21500
870
+ },
871
+ {
872
+ "epoch": 0.9969637920877328,
873
+ "grad_norm": 0.6297094821929932,
874
+ "learning_rate": 3.338393679853779e-05,
875
+ "loss": 0.8159,
876
+ "step": 22000
877
+ },
878
+ {
879
+ "epoch": 0.9969637920877328,
880
+ "eval_bleu": 38.4628,
881
+ "eval_chrf++": 64.9748,
882
+ "eval_gen_len": 27.7568,
883
+ "eval_loss": 0.7950631380081177,
884
+ "eval_runtime": 757.8415,
885
+ "eval_samples_per_second": 3.299,
886
+ "eval_spbleu": 52.6788,
887
+ "eval_steps_per_second": 0.66,
888
+ "eval_ter": 45.3491,
889
+ "step": 22000
890
+ },
891
+ {
892
+ "epoch": 1.0196220600897268,
893
+ "grad_norm": 0.6592913866043091,
894
+ "learning_rate": 3.300629899850455e-05,
895
+ "loss": 0.6601,
896
+ "step": 22500
897
+ },
898
+ {
899
+ "epoch": 1.0196220600897268,
900
+ "eval_bleu": 39.9752,
901
+ "eval_chrf++": 65.1542,
902
+ "eval_gen_len": 27.834,
903
+ "eval_loss": 0.810078501701355,
904
+ "eval_runtime": 742.5172,
905
+ "eval_samples_per_second": 3.367,
906
+ "eval_spbleu": 52.8176,
907
+ "eval_steps_per_second": 0.673,
908
+ "eval_ter": 45.4176,
909
+ "step": 22500
910
+ },
911
+ {
912
+ "epoch": 1.0422803280917208,
913
+ "grad_norm": 0.5827597379684448,
914
+ "learning_rate": 3.262866119847132e-05,
915
+ "loss": 0.6316,
916
+ "step": 23000
917
+ },
918
+ {
919
+ "epoch": 1.0422803280917208,
920
+ "eval_bleu": 43.5482,
921
+ "eval_chrf++": 65.346,
922
+ "eval_gen_len": 27.59,
923
+ "eval_loss": 0.814832329750061,
924
+ "eval_runtime": 736.0286,
925
+ "eval_samples_per_second": 3.397,
926
+ "eval_spbleu": 52.9228,
927
+ "eval_steps_per_second": 0.679,
928
+ "eval_ter": 45.2832,
929
+ "step": 23000
930
+ },
931
+ {
932
+ "epoch": 1.0649385960937146,
933
+ "grad_norm": 0.736733078956604,
934
+ "learning_rate": 3.225102339843809e-05,
935
+ "loss": 0.6365,
936
+ "step": 23500
937
+ },
938
+ {
939
+ "epoch": 1.0649385960937146,
940
+ "eval_bleu": 40.0573,
941
+ "eval_chrf++": 65.2248,
942
+ "eval_gen_len": 27.7468,
943
+ "eval_loss": 0.8096536993980408,
944
+ "eval_runtime": 762.5207,
945
+ "eval_samples_per_second": 3.279,
946
+ "eval_spbleu": 52.9015,
947
+ "eval_steps_per_second": 0.656,
948
+ "eval_ter": 45.1302,
949
+ "step": 23500
950
+ },
951
+ {
952
+ "epoch": 1.0875968640957085,
953
+ "grad_norm": 0.6646838188171387,
954
+ "learning_rate": 3.187338559840486e-05,
955
+ "loss": 0.6462,
956
+ "step": 24000
957
+ },
958
+ {
959
+ "epoch": 1.0875968640957085,
960
+ "eval_bleu": 41.7287,
961
+ "eval_chrf++": 65.476,
962
+ "eval_gen_len": 27.8052,
963
+ "eval_loss": 0.8067141771316528,
964
+ "eval_runtime": 764.719,
965
+ "eval_samples_per_second": 3.269,
966
+ "eval_spbleu": 53.0853,
967
+ "eval_steps_per_second": 0.654,
968
+ "eval_ter": 45.2621,
969
+ "step": 24000
970
+ },
971
+ {
972
+ "epoch": 1.1102551320977025,
973
+ "grad_norm": 0.7113286852836609,
974
+ "learning_rate": 3.149574779837163e-05,
975
+ "loss": 0.6383,
976
+ "step": 24500
977
+ },
978
+ {
979
+ "epoch": 1.1102551320977025,
980
+ "eval_bleu": 44.484,
981
+ "eval_chrf++": 65.5094,
982
+ "eval_gen_len": 27.502,
983
+ "eval_loss": 0.8042193055152893,
984
+ "eval_runtime": 761.5932,
985
+ "eval_samples_per_second": 3.283,
986
+ "eval_spbleu": 53.234,
987
+ "eval_steps_per_second": 0.657,
988
+ "eval_ter": 45.0591,
989
+ "step": 24500
990
+ },
991
+ {
992
+ "epoch": 1.1329134000996963,
993
+ "grad_norm": 0.6746016144752502,
994
+ "learning_rate": 3.1118109998338396e-05,
995
+ "loss": 0.6464,
996
+ "step": 25000
997
+ },
998
+ {
999
+ "epoch": 1.1329134000996963,
1000
+ "eval_bleu": 44.4968,
1001
+ "eval_chrf++": 65.4383,
1002
+ "eval_gen_len": 27.4832,
1003
+ "eval_loss": 0.8051723837852478,
1004
+ "eval_runtime": 762.6344,
1005
+ "eval_samples_per_second": 3.278,
1006
+ "eval_spbleu": 53.209,
1007
+ "eval_steps_per_second": 0.656,
1008
+ "eval_ter": 45.1672,
1009
+ "step": 25000
1010
+ },
1011
+ {
1012
+ "epoch": 1.1555716681016903,
1013
+ "grad_norm": 0.7094623446464539,
1014
+ "learning_rate": 3.074047219830516e-05,
1015
+ "loss": 0.6353,
1016
+ "step": 25500
1017
+ },
1018
+ {
1019
+ "epoch": 1.1555716681016903,
1020
+ "eval_bleu": 44.7381,
1021
+ "eval_chrf++": 65.7012,
1022
+ "eval_gen_len": 27.5904,
1023
+ "eval_loss": 0.8053088784217834,
1024
+ "eval_runtime": 766.039,
1025
+ "eval_samples_per_second": 3.264,
1026
+ "eval_spbleu": 53.2617,
1027
+ "eval_steps_per_second": 0.653,
1028
+ "eval_ter": 44.9852,
1029
+ "step": 25500
1030
+ },
1031
+ {
1032
+ "epoch": 1.1782299361036843,
1033
+ "grad_norm": 0.5892546772956848,
1034
+ "learning_rate": 3.0362834398271934e-05,
1035
+ "loss": 0.6483,
1036
+ "step": 26000
1037
+ },
1038
+ {
1039
+ "epoch": 1.1782299361036843,
1040
+ "eval_bleu": 44.1957,
1041
+ "eval_chrf++": 65.3258,
1042
+ "eval_gen_len": 27.6048,
1043
+ "eval_loss": 0.8034100532531738,
1044
+ "eval_runtime": 761.0595,
1045
+ "eval_samples_per_second": 3.285,
1046
+ "eval_spbleu": 52.8918,
1047
+ "eval_steps_per_second": 0.657,
1048
+ "eval_ter": 45.4572,
1049
+ "step": 26000
1050
+ },
1051
+ {
1052
+ "epoch": 1.200888204105678,
1053
+ "grad_norm": 0.744484543800354,
1054
+ "learning_rate": 2.9985196598238697e-05,
1055
+ "loss": 0.6382,
1056
+ "step": 26500
1057
+ },
1058
+ {
1059
+ "epoch": 1.200888204105678,
1060
+ "eval_bleu": 44.663,
1061
+ "eval_chrf++": 65.2386,
1062
+ "eval_gen_len": 27.4888,
1063
+ "eval_loss": 0.8034644722938538,
1064
+ "eval_runtime": 742.6066,
1065
+ "eval_samples_per_second": 3.367,
1066
+ "eval_spbleu": 52.8205,
1067
+ "eval_steps_per_second": 0.673,
1068
+ "eval_ter": 45.1144,
1069
+ "step": 26500
1070
+ },
1071
+ {
1072
+ "epoch": 1.223546472107672,
1073
+ "grad_norm": 0.6990212798118591,
1074
+ "learning_rate": 2.9607558798205465e-05,
1075
+ "loss": 0.6425,
1076
+ "step": 27000
1077
+ },
1078
+ {
1079
+ "epoch": 1.223546472107672,
1080
+ "eval_bleu": 40.3169,
1081
+ "eval_chrf++": 65.5055,
1082
+ "eval_gen_len": 27.7456,
1083
+ "eval_loss": 0.7975181341171265,
1084
+ "eval_runtime": 753.8226,
1085
+ "eval_samples_per_second": 3.316,
1086
+ "eval_spbleu": 53.4546,
1087
+ "eval_steps_per_second": 0.663,
1088
+ "eval_ter": 44.7295,
1089
+ "step": 27000
1090
+ },
1091
+ {
1092
+ "epoch": 1.246204740109666,
1093
+ "grad_norm": 0.7708460092544556,
1094
+ "learning_rate": 2.9229920998172232e-05,
1095
+ "loss": 0.6364,
1096
+ "step": 27500
1097
+ },
1098
+ {
1099
+ "epoch": 1.246204740109666,
1100
+ "eval_bleu": 39.0833,
1101
+ "eval_chrf++": 65.4463,
1102
+ "eval_gen_len": 27.9864,
1103
+ "eval_loss": 0.799001157283783,
1104
+ "eval_runtime": 765.257,
1105
+ "eval_samples_per_second": 3.267,
1106
+ "eval_spbleu": 53.3214,
1107
+ "eval_steps_per_second": 0.653,
1108
+ "eval_ter": 45.1513,
1109
+ "step": 27500
1110
+ },
1111
+ {
1112
+ "epoch": 1.2688630081116599,
1113
+ "grad_norm": 0.7766411900520325,
1114
+ "learning_rate": 2.8852283198139002e-05,
1115
+ "loss": 0.6311,
1116
+ "step": 28000
1117
+ },
1118
+ {
1119
+ "epoch": 1.2688630081116599,
1120
+ "eval_bleu": 44.6719,
1121
+ "eval_chrf++": 65.7788,
1122
+ "eval_gen_len": 27.5984,
1123
+ "eval_loss": 0.8024120330810547,
1124
+ "eval_runtime": 739.5078,
1125
+ "eval_samples_per_second": 3.381,
1126
+ "eval_spbleu": 53.4111,
1127
+ "eval_steps_per_second": 0.676,
1128
+ "eval_ter": 44.8429,
1129
+ "step": 28000
1130
+ },
1131
+ {
1132
+ "epoch": 1.2915212761136539,
1133
+ "grad_norm": 0.6488195657730103,
1134
+ "learning_rate": 2.847464539810577e-05,
1135
+ "loss": 0.6315,
1136
+ "step": 28500
1137
+ },
1138
+ {
1139
+ "epoch": 1.2915212761136539,
1140
+ "eval_bleu": 38.9676,
1141
+ "eval_chrf++": 65.2009,
1142
+ "eval_gen_len": 27.7544,
1143
+ "eval_loss": 0.799105167388916,
1144
+ "eval_runtime": 751.4345,
1145
+ "eval_samples_per_second": 3.327,
1146
+ "eval_spbleu": 53.0925,
1147
+ "eval_steps_per_second": 0.665,
1148
+ "eval_ter": 45.0802,
1149
+ "step": 28500
1150
+ },
1151
+ {
1152
+ "epoch": 1.3141795441156479,
1153
+ "grad_norm": 0.6332802176475525,
1154
+ "learning_rate": 2.809700759807254e-05,
1155
+ "loss": 0.6339,
1156
+ "step": 29000
1157
+ },
1158
+ {
1159
+ "epoch": 1.3141795441156479,
1160
+ "eval_bleu": 39.0276,
1161
+ "eval_chrf++": 65.2617,
1162
+ "eval_gen_len": 27.824,
1163
+ "eval_loss": 0.7974073886871338,
1164
+ "eval_runtime": 754.6502,
1165
+ "eval_samples_per_second": 3.313,
1166
+ "eval_spbleu": 53.046,
1167
+ "eval_steps_per_second": 0.663,
1168
+ "eval_ter": 45.2331,
1169
+ "step": 29000
1170
+ },
1171
+ {
1172
+ "epoch": 1.3368378121176416,
1173
+ "grad_norm": 0.5959407687187195,
1174
+ "learning_rate": 2.7719369798039307e-05,
1175
+ "loss": 0.6412,
1176
+ "step": 29500
1177
+ },
1178
+ {
1179
+ "epoch": 1.3368378121176416,
1180
+ "eval_bleu": 40.1118,
1181
+ "eval_chrf++": 65.4814,
1182
+ "eval_gen_len": 27.8892,
1183
+ "eval_loss": 0.7944240570068359,
1184
+ "eval_runtime": 752.0202,
1185
+ "eval_samples_per_second": 3.324,
1186
+ "eval_spbleu": 53.4305,
1187
+ "eval_steps_per_second": 0.665,
1188
+ "eval_ter": 45.0327,
1189
+ "step": 29500
1190
+ },
1191
+ {
1192
+ "epoch": 1.3594960801196356,
1193
+ "grad_norm": 0.6927244067192078,
1194
+ "learning_rate": 2.7341731998006075e-05,
1195
+ "loss": 0.6354,
1196
+ "step": 30000
1197
+ },
1198
+ {
1199
+ "epoch": 1.3594960801196356,
1200
+ "eval_bleu": 41.7725,
1201
+ "eval_chrf++": 65.7724,
1202
+ "eval_gen_len": 27.7264,
1203
+ "eval_loss": 0.7973920702934265,
1204
+ "eval_runtime": 742.892,
1205
+ "eval_samples_per_second": 3.365,
1206
+ "eval_spbleu": 53.752,
1207
+ "eval_steps_per_second": 0.673,
1208
+ "eval_ter": 44.4975,
1209
+ "step": 30000
1210
+ },
1211
+ {
1212
+ "epoch": 1.3821543481216296,
1213
+ "grad_norm": 0.6661298871040344,
1214
+ "learning_rate": 2.696409419797284e-05,
1215
+ "loss": 0.6294,
1216
+ "step": 30500
1217
+ },
1218
+ {
1219
+ "epoch": 1.3821543481216296,
1220
+ "eval_bleu": 39.0417,
1221
+ "eval_chrf++": 65.4211,
1222
+ "eval_gen_len": 27.8648,
1223
+ "eval_loss": 0.795570969581604,
1224
+ "eval_runtime": 749.9034,
1225
+ "eval_samples_per_second": 3.334,
1226
+ "eval_spbleu": 53.378,
1227
+ "eval_steps_per_second": 0.667,
1228
+ "eval_ter": 45.0802,
1229
+ "step": 30500
1230
+ },
1231
+ {
1232
+ "epoch": 1.4048126161236234,
1233
+ "grad_norm": 0.6107171773910522,
1234
+ "learning_rate": 2.658645639793961e-05,
1235
+ "loss": 0.636,
1236
+ "step": 31000
1237
+ },
1238
+ {
1239
+ "epoch": 1.4048126161236234,
1240
+ "eval_bleu": 39.9268,
1241
+ "eval_chrf++": 65.47,
1242
+ "eval_gen_len": 27.682,
1243
+ "eval_loss": 0.7938565611839294,
1244
+ "eval_runtime": 743.0863,
1245
+ "eval_samples_per_second": 3.364,
1246
+ "eval_spbleu": 53.3727,
1247
+ "eval_steps_per_second": 0.673,
1248
+ "eval_ter": 44.5871,
1249
+ "step": 31000
1250
+ },
1251
+ {
1252
+ "epoch": 1.4274708841256174,
1253
+ "grad_norm": 0.6050147414207458,
1254
+ "learning_rate": 2.6208818597906376e-05,
1255
+ "loss": 0.6385,
1256
+ "step": 31500
1257
+ },
1258
+ {
1259
+ "epoch": 1.4274708841256174,
1260
+ "eval_bleu": 40.4175,
1261
+ "eval_chrf++": 65.7235,
1262
+ "eval_gen_len": 27.8052,
1263
+ "eval_loss": 0.7914307713508606,
1264
+ "eval_runtime": 748.359,
1265
+ "eval_samples_per_second": 3.341,
1266
+ "eval_spbleu": 53.722,
1267
+ "eval_steps_per_second": 0.668,
1268
+ "eval_ter": 44.6003,
1269
+ "step": 31500
1270
+ },
1271
+ {
1272
+ "epoch": 1.4501291521276114,
1273
+ "grad_norm": 0.8934792280197144,
1274
+ "learning_rate": 2.5831180797873143e-05,
1275
+ "loss": 0.6392,
1276
+ "step": 32000
1277
+ },
1278
+ {
1279
+ "epoch": 1.4501291521276114,
1280
+ "eval_bleu": 40.6796,
1281
+ "eval_chrf++": 65.8163,
1282
+ "eval_gen_len": 27.7424,
1283
+ "eval_loss": 0.7927303910255432,
1284
+ "eval_runtime": 748.6849,
1285
+ "eval_samples_per_second": 3.339,
1286
+ "eval_spbleu": 53.7139,
1287
+ "eval_steps_per_second": 0.668,
1288
+ "eval_ter": 44.4184,
1289
+ "step": 32000
1290
+ },
1291
+ {
1292
+ "epoch": 1.4727874201296052,
1293
+ "grad_norm": 0.6742972731590271,
1294
+ "learning_rate": 2.5453542997839914e-05,
1295
+ "loss": 0.6364,
1296
+ "step": 32500
1297
+ },
1298
+ {
1299
+ "epoch": 1.4727874201296052,
1300
+ "eval_bleu": 40.2137,
1301
+ "eval_chrf++": 65.6498,
1302
+ "eval_gen_len": 27.7408,
1303
+ "eval_loss": 0.7901710867881775,
1304
+ "eval_runtime": 749.8815,
1305
+ "eval_samples_per_second": 3.334,
1306
+ "eval_spbleu": 53.6947,
1307
+ "eval_steps_per_second": 0.667,
1308
+ "eval_ter": 44.7031,
1309
+ "step": 32500
1310
+ },
1311
+ {
1312
+ "epoch": 1.4954456881315992,
1313
+ "grad_norm": 0.6159557104110718,
1314
+ "learning_rate": 2.507590519780668e-05,
1315
+ "loss": 0.6352,
1316
+ "step": 33000
1317
+ },
1318
+ {
1319
+ "epoch": 1.4954456881315992,
1320
+ "eval_bleu": 41.264,
1321
+ "eval_chrf++": 65.7523,
1322
+ "eval_gen_len": 27.8552,
1323
+ "eval_loss": 0.7894487380981445,
1324
+ "eval_runtime": 748.2109,
1325
+ "eval_samples_per_second": 3.341,
1326
+ "eval_spbleu": 53.6724,
1327
+ "eval_steps_per_second": 0.668,
1328
+ "eval_ter": 44.8244,
1329
+ "step": 33000
1330
+ },
1331
+ {
1332
+ "epoch": 1.5181039561335932,
1333
+ "grad_norm": 0.8539830446243286,
1334
+ "learning_rate": 2.469826739777345e-05,
1335
+ "loss": 0.6234,
1336
+ "step": 33500
1337
+ },
1338
+ {
1339
+ "epoch": 1.5181039561335932,
1340
+ "eval_bleu": 40.1811,
1341
+ "eval_chrf++": 65.8305,
1342
+ "eval_gen_len": 27.8484,
1343
+ "eval_loss": 0.7886767983436584,
1344
+ "eval_runtime": 745.5767,
1345
+ "eval_samples_per_second": 3.353,
1346
+ "eval_spbleu": 53.7788,
1347
+ "eval_steps_per_second": 0.671,
1348
+ "eval_ter": 44.5739,
1349
+ "step": 33500
1350
+ },
1351
+ {
1352
+ "epoch": 1.540762224135587,
1353
+ "grad_norm": 0.7526208758354187,
1354
+ "learning_rate": 2.4320629597740216e-05,
1355
+ "loss": 0.6411,
1356
+ "step": 34000
1357
+ },
1358
+ {
1359
+ "epoch": 1.540762224135587,
1360
+ "eval_bleu": 42.6857,
1361
+ "eval_chrf++": 65.9528,
1362
+ "eval_gen_len": 27.6804,
1363
+ "eval_loss": 0.7847135663032532,
1364
+ "eval_runtime": 741.0515,
1365
+ "eval_samples_per_second": 3.374,
1366
+ "eval_spbleu": 53.8997,
1367
+ "eval_steps_per_second": 0.675,
1368
+ "eval_ter": 44.3024,
1369
+ "step": 34000
1370
+ },
1371
+ {
1372
+ "epoch": 1.5634204921375812,
1373
+ "grad_norm": 0.8738523125648499,
1374
+ "learning_rate": 2.3942991797706986e-05,
1375
+ "loss": 0.63,
1376
+ "step": 34500
1377
+ },
1378
+ {
1379
+ "epoch": 1.5634204921375812,
1380
+ "eval_bleu": 45.4639,
1381
+ "eval_chrf++": 66.1251,
1382
+ "eval_gen_len": 27.5308,
1383
+ "eval_loss": 0.781486451625824,
1384
+ "eval_runtime": 736.5596,
1385
+ "eval_samples_per_second": 3.394,
1386
+ "eval_spbleu": 53.9656,
1387
+ "eval_steps_per_second": 0.679,
1388
+ "eval_ter": 44.334,
1389
+ "step": 34500
1390
+ },
1391
+ {
1392
+ "epoch": 1.586078760139575,
1393
+ "grad_norm": 0.7074981927871704,
1394
+ "learning_rate": 2.3565353997673753e-05,
1395
+ "loss": 0.6265,
1396
+ "step": 35000
1397
+ },
1398
+ {
1399
+ "epoch": 1.586078760139575,
1400
+ "eval_bleu": 44.9751,
1401
+ "eval_chrf++": 65.9036,
1402
+ "eval_gen_len": 27.6124,
1403
+ "eval_loss": 0.7836451530456543,
1404
+ "eval_runtime": 746.2524,
1405
+ "eval_samples_per_second": 3.35,
1406
+ "eval_spbleu": 53.5997,
1407
+ "eval_steps_per_second": 0.67,
1408
+ "eval_ter": 44.8059,
1409
+ "step": 35000
1410
+ },
1411
+ {
1412
+ "epoch": 1.6087370281415687,
1413
+ "grad_norm": 0.7853338718414307,
1414
+ "learning_rate": 2.318771619764052e-05,
1415
+ "loss": 0.6202,
1416
+ "step": 35500
1417
+ },
1418
+ {
1419
+ "epoch": 1.6087370281415687,
1420
+ "eval_bleu": 45.0486,
1421
+ "eval_chrf++": 66.1202,
1422
+ "eval_gen_len": 27.6572,
1423
+ "eval_loss": 0.7826634049415588,
1424
+ "eval_runtime": 750.4591,
1425
+ "eval_samples_per_second": 3.331,
1426
+ "eval_spbleu": 53.9114,
1427
+ "eval_steps_per_second": 0.666,
1428
+ "eval_ter": 44.7242,
1429
+ "step": 35500
1430
+ },
1431
+ {
1432
+ "epoch": 1.631395296143563,
1433
+ "grad_norm": 0.7461378574371338,
1434
+ "learning_rate": 2.2810078397607288e-05,
1435
+ "loss": 0.6385,
1436
+ "step": 36000
1437
+ },
1438
+ {
1439
+ "epoch": 1.631395296143563,
1440
+ "eval_bleu": 41.8569,
1441
+ "eval_chrf++": 65.9455,
1442
+ "eval_gen_len": 27.8612,
1443
+ "eval_loss": 0.7859405279159546,
1444
+ "eval_runtime": 753.5356,
1445
+ "eval_samples_per_second": 3.318,
1446
+ "eval_spbleu": 53.9218,
1447
+ "eval_steps_per_second": 0.664,
1448
+ "eval_ter": 44.5897,
1449
+ "step": 36000
1450
+ },
1451
+ {
1452
+ "epoch": 1.6540535641455567,
1453
+ "grad_norm": 0.5514925122261047,
1454
+ "learning_rate": 2.2432440597574055e-05,
1455
+ "loss": 0.6269,
1456
+ "step": 36500
1457
+ },
1458
+ {
1459
+ "epoch": 1.6540535641455567,
1460
+ "eval_bleu": 43.6861,
1461
+ "eval_chrf++": 65.9551,
1462
+ "eval_gen_len": 27.5088,
1463
+ "eval_loss": 0.7851018905639648,
1464
+ "eval_runtime": 744.0917,
1465
+ "eval_samples_per_second": 3.36,
1466
+ "eval_spbleu": 53.792,
1467
+ "eval_steps_per_second": 0.672,
1468
+ "eval_ter": 44.3419,
1469
+ "step": 36500
1470
+ },
1471
+ {
1472
+ "epoch": 1.6767118321475505,
1473
+ "grad_norm": 0.6642000675201416,
1474
+ "learning_rate": 2.2054802797540825e-05,
1475
+ "loss": 0.6301,
1476
+ "step": 37000
1477
+ },
1478
+ {
1479
+ "epoch": 1.6767118321475505,
1480
+ "eval_bleu": 46.0896,
1481
+ "eval_chrf++": 66.164,
1482
+ "eval_gen_len": 27.426,
1483
+ "eval_loss": 0.7796212434768677,
1484
+ "eval_runtime": 740.9337,
1485
+ "eval_samples_per_second": 3.374,
1486
+ "eval_spbleu": 54.0105,
1487
+ "eval_steps_per_second": 0.675,
1488
+ "eval_ter": 44.1494,
1489
+ "step": 37000
1490
+ },
1491
+ {
1492
+ "epoch": 1.6993701001495447,
1493
+ "grad_norm": 0.8100460171699524,
1494
+ "learning_rate": 2.167716499750759e-05,
1495
+ "loss": 0.6213,
1496
+ "step": 37500
1497
+ },
1498
+ {
1499
+ "epoch": 1.6993701001495447,
1500
+ "eval_bleu": 45.5601,
1501
+ "eval_chrf++": 66.0823,
1502
+ "eval_gen_len": 27.5128,
1503
+ "eval_loss": 0.7815007567405701,
1504
+ "eval_runtime": 733.2938,
1505
+ "eval_samples_per_second": 3.409,
1506
+ "eval_spbleu": 53.9922,
1507
+ "eval_steps_per_second": 0.682,
1508
+ "eval_ter": 44.5133,
1509
+ "step": 37500
1510
+ },
1511
+ {
1512
+ "epoch": 1.7220283681515385,
1513
+ "grad_norm": 0.6142133474349976,
1514
+ "learning_rate": 2.129952719747436e-05,
1515
+ "loss": 0.623,
1516
+ "step": 38000
1517
+ },
1518
+ {
1519
+ "epoch": 1.7220283681515385,
1520
+ "eval_bleu": 45.0364,
1521
+ "eval_chrf++": 66.1218,
1522
+ "eval_gen_len": 27.5352,
1523
+ "eval_loss": 0.7782283425331116,
1524
+ "eval_runtime": 736.7203,
1525
+ "eval_samples_per_second": 3.393,
1526
+ "eval_spbleu": 54.0624,
1527
+ "eval_steps_per_second": 0.679,
1528
+ "eval_ter": 44.3314,
1529
+ "step": 38000
1530
+ },
1531
+ {
1532
+ "epoch": 1.7446866361535325,
1533
+ "grad_norm": 0.8021434545516968,
1534
+ "learning_rate": 2.0921889397441127e-05,
1535
+ "loss": 0.6269,
1536
+ "step": 38500
1537
+ },
1538
+ {
1539
+ "epoch": 1.7446866361535325,
1540
+ "eval_bleu": 41.6796,
1541
+ "eval_chrf++": 66.0402,
1542
+ "eval_gen_len": 27.7448,
1543
+ "eval_loss": 0.7799319624900818,
1544
+ "eval_runtime": 738.6103,
1545
+ "eval_samples_per_second": 3.385,
1546
+ "eval_spbleu": 54.1452,
1547
+ "eval_steps_per_second": 0.677,
1548
+ "eval_ter": 44.1653,
1549
+ "step": 38500
1550
+ },
1551
+ {
1552
+ "epoch": 1.7673449041555265,
1553
+ "grad_norm": 0.6603755354881287,
1554
+ "learning_rate": 2.0544251597407894e-05,
1555
+ "loss": 0.6339,
1556
+ "step": 39000
1557
+ },
1558
+ {
1559
+ "epoch": 1.7673449041555265,
1560
+ "eval_bleu": 46.523,
1561
+ "eval_chrf++": 66.3925,
1562
+ "eval_gen_len": 27.5112,
1563
+ "eval_loss": 0.7813342809677124,
1564
+ "eval_runtime": 728.2882,
1565
+ "eval_samples_per_second": 3.433,
1566
+ "eval_spbleu": 54.2461,
1567
+ "eval_steps_per_second": 0.687,
1568
+ "eval_ter": 44.0071,
1569
+ "step": 39000
1570
+ },
1571
+ {
1572
+ "epoch": 1.7900031721575203,
1573
+ "grad_norm": 0.6977267861366272,
1574
+ "learning_rate": 2.016661379737466e-05,
1575
+ "loss": 0.621,
1576
+ "step": 39500
1577
+ },
1578
+ {
1579
+ "epoch": 1.7900031721575203,
1580
+ "eval_bleu": 43.8812,
1581
+ "eval_chrf++": 66.186,
1582
+ "eval_gen_len": 27.6576,
1583
+ "eval_loss": 0.7753216028213501,
1584
+ "eval_runtime": 736.3388,
1585
+ "eval_samples_per_second": 3.395,
1586
+ "eval_spbleu": 54.244,
1587
+ "eval_steps_per_second": 0.679,
1588
+ "eval_ter": 44.3445,
1589
+ "step": 39500
1590
+ },
1591
+ {
1592
+ "epoch": 1.8126614401595142,
1593
+ "grad_norm": 0.7790645956993103,
1594
+ "learning_rate": 1.978897599734143e-05,
1595
+ "loss": 0.6278,
1596
+ "step": 40000
1597
+ },
1598
+ {
1599
+ "epoch": 1.8126614401595142,
1600
+ "eval_bleu": 46.7458,
1601
+ "eval_chrf++": 66.4123,
1602
+ "eval_gen_len": 27.4892,
1603
+ "eval_loss": 0.7777643799781799,
1604
+ "eval_runtime": 730.5316,
1605
+ "eval_samples_per_second": 3.422,
1606
+ "eval_spbleu": 54.4016,
1607
+ "eval_steps_per_second": 0.684,
1608
+ "eval_ter": 43.9702,
1609
+ "step": 40000
1610
+ },
1611
+ {
1612
+ "epoch": 1.8353197081615082,
1613
+ "grad_norm": 0.6901569366455078,
1614
+ "learning_rate": 1.94113381973082e-05,
1615
+ "loss": 0.6221,
1616
+ "step": 40500
1617
+ },
1618
+ {
1619
+ "epoch": 1.8353197081615082,
1620
+ "eval_bleu": 45.0544,
1621
+ "eval_chrf++": 66.3725,
1622
+ "eval_gen_len": 27.598,
1623
+ "eval_loss": 0.7787633538246155,
1624
+ "eval_runtime": 734.485,
1625
+ "eval_samples_per_second": 3.404,
1626
+ "eval_spbleu": 54.419,
1627
+ "eval_steps_per_second": 0.681,
1628
+ "eval_ter": 43.9201,
1629
+ "step": 40500
1630
+ },
1631
+ {
1632
+ "epoch": 1.857977976163502,
1633
+ "grad_norm": 0.7807871103286743,
1634
+ "learning_rate": 1.9033700397274966e-05,
1635
+ "loss": 0.6209,
1636
+ "step": 41000
1637
+ },
1638
+ {
1639
+ "epoch": 1.857977976163502,
1640
+ "eval_bleu": 44.5498,
1641
+ "eval_chrf++": 66.3741,
1642
+ "eval_gen_len": 27.5916,
1643
+ "eval_loss": 0.7768906354904175,
1644
+ "eval_runtime": 737.3971,
1645
+ "eval_samples_per_second": 3.39,
1646
+ "eval_spbleu": 54.5028,
1647
+ "eval_steps_per_second": 0.678,
1648
+ "eval_ter": 43.9728,
1649
+ "step": 41000
1650
+ },
1651
+ {
1652
+ "epoch": 1.880636244165496,
1653
+ "grad_norm": 0.8082613945007324,
1654
+ "learning_rate": 1.8656062597241737e-05,
1655
+ "loss": 0.6267,
1656
+ "step": 41500
1657
+ },
1658
+ {
1659
+ "epoch": 1.880636244165496,
1660
+ "eval_bleu": 45.3502,
1661
+ "eval_chrf++": 66.5334,
1662
+ "eval_gen_len": 27.5344,
1663
+ "eval_loss": 0.7741044759750366,
1664
+ "eval_runtime": 731.6398,
1665
+ "eval_samples_per_second": 3.417,
1666
+ "eval_spbleu": 54.4958,
1667
+ "eval_steps_per_second": 0.683,
1668
+ "eval_ter": 43.8726,
1669
+ "step": 41500
1670
+ },
1671
+ {
1672
+ "epoch": 1.90329451216749,
1673
+ "grad_norm": 0.5612310171127319,
1674
+ "learning_rate": 1.82784247972085e-05,
1675
+ "loss": 0.625,
1676
+ "step": 42000
1677
+ },
1678
+ {
1679
+ "epoch": 1.90329451216749,
1680
+ "eval_bleu": 45.4662,
1681
+ "eval_chrf++": 66.6858,
1682
+ "eval_gen_len": 27.5552,
1683
+ "eval_loss": 0.7751156687736511,
1684
+ "eval_runtime": 734.022,
1685
+ "eval_samples_per_second": 3.406,
1686
+ "eval_spbleu": 54.7854,
1687
+ "eval_steps_per_second": 0.681,
1688
+ "eval_ter": 43.5404,
1689
+ "step": 42000
1690
+ },
1691
+ {
1692
+ "epoch": 1.9259527801694838,
1693
+ "grad_norm": 0.7477275133132935,
1694
+ "learning_rate": 1.790078699717527e-05,
1695
+ "loss": 0.6268,
1696
+ "step": 42500
1697
+ },
1698
+ {
1699
+ "epoch": 1.9259527801694838,
1700
+ "eval_bleu": 43.7231,
1701
+ "eval_chrf++": 66.4796,
1702
+ "eval_gen_len": 27.71,
1703
+ "eval_loss": 0.7729161977767944,
1704
+ "eval_runtime": 736.8783,
1705
+ "eval_samples_per_second": 3.393,
1706
+ "eval_spbleu": 54.5524,
1707
+ "eval_steps_per_second": 0.679,
1708
+ "eval_ter": 43.8383,
1709
+ "step": 42500
1710
+ },
1711
+ {
1712
+ "epoch": 1.9486110481714778,
1713
+ "grad_norm": 0.7059822678565979,
1714
+ "learning_rate": 1.752314919714204e-05,
1715
+ "loss": 0.6263,
1716
+ "step": 43000
1717
+ },
1718
+ {
1719
+ "epoch": 1.9486110481714778,
1720
+ "eval_bleu": 45.1162,
1721
+ "eval_chrf++": 66.5293,
1722
+ "eval_gen_len": 27.5624,
1723
+ "eval_loss": 0.771515429019928,
1724
+ "eval_runtime": 732.1583,
1725
+ "eval_samples_per_second": 3.415,
1726
+ "eval_spbleu": 54.5958,
1727
+ "eval_steps_per_second": 0.683,
1728
+ "eval_ter": 43.6564,
1729
+ "step": 43000
1730
+ },
1731
+ {
1732
+ "epoch": 1.9712693161734718,
1733
+ "grad_norm": 0.7907470464706421,
1734
+ "learning_rate": 1.7145511397108806e-05,
1735
+ "loss": 0.6178,
1736
+ "step": 43500
1737
+ },
1738
+ {
1739
+ "epoch": 1.9712693161734718,
1740
+ "eval_bleu": 44.3099,
1741
+ "eval_chrf++": 66.5724,
1742
+ "eval_gen_len": 27.632,
1743
+ "eval_loss": 0.7728075385093689,
1744
+ "eval_runtime": 733.6074,
1745
+ "eval_samples_per_second": 3.408,
1746
+ "eval_spbleu": 54.6752,
1747
+ "eval_steps_per_second": 0.682,
1748
+ "eval_ter": 43.725,
1749
+ "step": 43500
1750
+ },
1751
+ {
1752
+ "epoch": 1.9939275841754656,
1753
+ "grad_norm": 0.7742732763290405,
1754
+ "learning_rate": 1.6767873597075573e-05,
1755
+ "loss": 0.609,
1756
+ "step": 44000
1757
+ },
1758
+ {
1759
+ "epoch": 1.9939275841754656,
1760
+ "eval_bleu": 46.4711,
1761
+ "eval_chrf++": 66.7721,
1762
+ "eval_gen_len": 27.4864,
1763
+ "eval_loss": 0.7715900540351868,
1764
+ "eval_runtime": 726.1293,
1765
+ "eval_samples_per_second": 3.443,
1766
+ "eval_spbleu": 54.9332,
1767
+ "eval_steps_per_second": 0.689,
1768
+ "eval_ter": 43.3822,
1769
+ "step": 44000
1770
+ },
1771
+ {
1772
+ "epoch": 2.0165858521774593,
1773
+ "grad_norm": 0.7575493454933167,
1774
+ "learning_rate": 1.639023579704234e-05,
1775
+ "loss": 0.5124,
1776
+ "step": 44500
1777
+ },
1778
+ {
1779
+ "epoch": 2.0165858521774593,
1780
+ "eval_bleu": 45.9215,
1781
+ "eval_chrf++": 66.6482,
1782
+ "eval_gen_len": 27.622,
1783
+ "eval_loss": 0.8128123879432678,
1784
+ "eval_runtime": 732.4537,
1785
+ "eval_samples_per_second": 3.413,
1786
+ "eval_spbleu": 54.7846,
1787
+ "eval_steps_per_second": 0.683,
1788
+ "eval_ter": 43.7355,
1789
+ "step": 44500
1790
+ },
1791
+ {
1792
+ "epoch": 2.0392441201794536,
1793
+ "grad_norm": 0.732072114944458,
1794
+ "learning_rate": 1.601259799700911e-05,
1795
+ "loss": 0.4683,
1796
+ "step": 45000
1797
+ },
1798
+ {
1799
+ "epoch": 2.0392441201794536,
1800
+ "eval_bleu": 45.6558,
1801
+ "eval_chrf++": 66.455,
1802
+ "eval_gen_len": 27.5796,
1803
+ "eval_loss": 0.8135092258453369,
1804
+ "eval_runtime": 732.47,
1805
+ "eval_samples_per_second": 3.413,
1806
+ "eval_spbleu": 54.5066,
1807
+ "eval_steps_per_second": 0.683,
1808
+ "eval_ter": 44.0202,
1809
+ "step": 45000
1810
+ },
1811
+ {
1812
+ "epoch": 2.0619023881814473,
1813
+ "grad_norm": 0.7785657644271851,
1814
+ "learning_rate": 1.5634960196975878e-05,
1815
+ "loss": 0.4632,
1816
+ "step": 45500
1817
+ },
1818
+ {
1819
+ "epoch": 2.0619023881814473,
1820
+ "eval_bleu": 46.9546,
1821
+ "eval_chrf++": 66.5237,
1822
+ "eval_gen_len": 27.482,
1823
+ "eval_loss": 0.81900554895401,
1824
+ "eval_runtime": 734.696,
1825
+ "eval_samples_per_second": 3.403,
1826
+ "eval_spbleu": 54.4887,
1827
+ "eval_steps_per_second": 0.681,
1828
+ "eval_ter": 43.7988,
1829
+ "step": 45500
1830
+ },
1831
+ {
1832
+ "epoch": 2.0845606561834416,
1833
+ "grad_norm": 1.1765786409378052,
1834
+ "learning_rate": 1.5257322396942645e-05,
1835
+ "loss": 0.4696,
1836
+ "step": 46000
1837
+ },
1838
+ {
1839
+ "epoch": 2.0845606561834416,
1840
+ "eval_bleu": 47.083,
1841
+ "eval_chrf++": 66.8275,
1842
+ "eval_gen_len": 27.5328,
1843
+ "eval_loss": 0.8155868053436279,
1844
+ "eval_runtime": 736.5203,
1845
+ "eval_samples_per_second": 3.394,
1846
+ "eval_spbleu": 54.892,
1847
+ "eval_steps_per_second": 0.679,
1848
+ "eval_ter": 43.456,
1849
+ "step": 46000
1850
+ },
1851
+ {
1852
+ "epoch": 2.1072189241854353,
1853
+ "grad_norm": 0.8067004084587097,
1854
+ "learning_rate": 1.4879684596909412e-05,
1855
+ "loss": 0.4635,
1856
+ "step": 46500
1857
+ },
1858
+ {
1859
+ "epoch": 2.1072189241854353,
1860
+ "eval_bleu": 46.6541,
1861
+ "eval_chrf++": 66.7448,
1862
+ "eval_gen_len": 27.5476,
1863
+ "eval_loss": 0.8161126971244812,
1864
+ "eval_runtime": 740.7767,
1865
+ "eval_samples_per_second": 3.375,
1866
+ "eval_spbleu": 54.8934,
1867
+ "eval_steps_per_second": 0.675,
1868
+ "eval_ter": 43.3189,
1869
+ "step": 46500
1870
+ },
1871
+ {
1872
+ "epoch": 2.129877192187429,
1873
+ "grad_norm": 0.9044099450111389,
1874
+ "learning_rate": 1.4502046796876181e-05,
1875
+ "loss": 0.4725,
1876
+ "step": 47000
1877
+ },
1878
+ {
1879
+ "epoch": 2.129877192187429,
1880
+ "eval_bleu": 47.2452,
1881
+ "eval_chrf++": 66.8326,
1882
+ "eval_gen_len": 27.558,
1883
+ "eval_loss": 0.8134703040122986,
1884
+ "eval_runtime": 741.2625,
1885
+ "eval_samples_per_second": 3.373,
1886
+ "eval_spbleu": 54.9705,
1887
+ "eval_steps_per_second": 0.675,
1888
+ "eval_ter": 43.3216,
1889
+ "step": 47000
1890
+ },
1891
+ {
1892
+ "epoch": 2.1525354601894233,
1893
+ "grad_norm": 0.8520795106887817,
1894
+ "learning_rate": 1.412440899684295e-05,
1895
+ "loss": 0.4727,
1896
+ "step": 47500
1897
+ },
1898
+ {
1899
+ "epoch": 2.1525354601894233,
1900
+ "eval_bleu": 46.5313,
1901
+ "eval_chrf++": 66.6736,
1902
+ "eval_gen_len": 27.5296,
1903
+ "eval_loss": 0.820831835269928,
1904
+ "eval_runtime": 742.5016,
1905
+ "eval_samples_per_second": 3.367,
1906
+ "eval_spbleu": 54.5714,
1907
+ "eval_steps_per_second": 0.673,
1908
+ "eval_ter": 43.6722,
1909
+ "step": 47500
1910
+ },
1911
+ {
1912
+ "epoch": 2.175193728191417,
1913
+ "grad_norm": 0.7529011964797974,
1914
+ "learning_rate": 1.3746771196809716e-05,
1915
+ "loss": 0.4736,
1916
+ "step": 48000
1917
+ },
1918
+ {
1919
+ "epoch": 2.175193728191417,
1920
+ "eval_bleu": 46.3524,
1921
+ "eval_chrf++": 66.7416,
1922
+ "eval_gen_len": 27.5684,
1923
+ "eval_loss": 0.8130167126655579,
1924
+ "eval_runtime": 730.5257,
1925
+ "eval_samples_per_second": 3.422,
1926
+ "eval_spbleu": 54.8088,
1927
+ "eval_steps_per_second": 0.684,
1928
+ "eval_ter": 43.4718,
1929
+ "step": 48000
1930
+ },
1931
+ {
1932
+ "epoch": 2.197851996193411,
1933
+ "grad_norm": 0.7033498883247375,
1934
+ "learning_rate": 1.3369133396776484e-05,
1935
+ "loss": 0.471,
1936
+ "step": 48500
1937
+ },
1938
+ {
1939
+ "epoch": 2.197851996193411,
1940
+ "eval_bleu": 46.2188,
1941
+ "eval_chrf++": 66.702,
1942
+ "eval_gen_len": 27.6192,
1943
+ "eval_loss": 0.8164393305778503,
1944
+ "eval_runtime": 732.5336,
1945
+ "eval_samples_per_second": 3.413,
1946
+ "eval_spbleu": 54.6656,
1947
+ "eval_steps_per_second": 0.683,
1948
+ "eval_ter": 43.6511,
1949
+ "step": 48500
1950
+ },
1951
+ {
1952
+ "epoch": 2.220510264195405,
1953
+ "grad_norm": 0.7331113815307617,
1954
+ "learning_rate": 1.2991495596743253e-05,
1955
+ "loss": 0.4712,
1956
+ "step": 49000
1957
+ },
1958
+ {
1959
+ "epoch": 2.220510264195405,
1960
+ "eval_bleu": 47.0435,
1961
+ "eval_chrf++": 66.6968,
1962
+ "eval_gen_len": 27.4924,
1963
+ "eval_loss": 0.81520676612854,
1964
+ "eval_runtime": 729.4123,
1965
+ "eval_samples_per_second": 3.427,
1966
+ "eval_spbleu": 54.6842,
1967
+ "eval_steps_per_second": 0.685,
1968
+ "eval_ter": 43.4376,
1969
+ "step": 49000
1970
+ },
1971
+ {
1972
+ "epoch": 2.243168532197399,
1973
+ "grad_norm": 0.6714054346084595,
1974
+ "learning_rate": 1.261385779671002e-05,
1975
+ "loss": 0.4741,
1976
+ "step": 49500
1977
+ },
1978
+ {
1979
+ "epoch": 2.243168532197399,
1980
+ "eval_bleu": 47.2441,
1981
+ "eval_chrf++": 66.8706,
1982
+ "eval_gen_len": 27.4916,
1983
+ "eval_loss": 0.8153889775276184,
1984
+ "eval_runtime": 727.829,
1985
+ "eval_samples_per_second": 3.435,
1986
+ "eval_spbleu": 54.9194,
1987
+ "eval_steps_per_second": 0.687,
1988
+ "eval_ter": 43.3374,
1989
+ "step": 49500
1990
+ },
1991
+ {
1992
+ "epoch": 2.2658268001993926,
1993
+ "grad_norm": 0.7230417132377625,
1994
+ "learning_rate": 1.2236219996676788e-05,
1995
+ "loss": 0.4723,
1996
+ "step": 50000
1997
+ },
1998
+ {
1999
+ "epoch": 2.2658268001993926,
2000
+ "eval_bleu": 47.0146,
2001
+ "eval_chrf++": 66.9999,
2002
+ "eval_gen_len": 27.496,
2003
+ "eval_loss": 0.8151711225509644,
2004
+ "eval_runtime": 725.7357,
2005
+ "eval_samples_per_second": 3.445,
2006
+ "eval_spbleu": 55.1257,
2007
+ "eval_steps_per_second": 0.689,
2008
+ "eval_ter": 43.0605,
2009
+ "step": 50000
2010
+ },
2011
+ {
2012
+ "epoch": 2.288485068201387,
2013
+ "grad_norm": 0.7548694014549255,
2014
+ "learning_rate": 1.1858582196643555e-05,
2015
+ "loss": 0.4736,
2016
+ "step": 50500
2017
+ },
2018
+ {
2019
+ "epoch": 2.288485068201387,
2020
+ "eval_bleu": 47.3114,
2021
+ "eval_chrf++": 67.059,
2022
+ "eval_gen_len": 27.5484,
2023
+ "eval_loss": 0.8111055493354797,
2024
+ "eval_runtime": 730.4241,
2025
+ "eval_samples_per_second": 3.423,
2026
+ "eval_spbleu": 55.1401,
2027
+ "eval_steps_per_second": 0.685,
2028
+ "eval_ter": 43.2688,
2029
+ "step": 50500
2030
+ },
2031
+ {
2032
+ "epoch": 2.3111433362033806,
2033
+ "grad_norm": 0.6914283037185669,
2034
+ "learning_rate": 1.1480944396610324e-05,
2035
+ "loss": 0.4673,
2036
+ "step": 51000
2037
+ },
2038
+ {
2039
+ "epoch": 2.3111433362033806,
2040
+ "eval_bleu": 47.0659,
2041
+ "eval_chrf++": 66.8804,
2042
+ "eval_gen_len": 27.5096,
2043
+ "eval_loss": 0.8131672739982605,
2044
+ "eval_runtime": 728.8458,
2045
+ "eval_samples_per_second": 3.43,
2046
+ "eval_spbleu": 54.9036,
2047
+ "eval_steps_per_second": 0.686,
2048
+ "eval_ter": 43.3585,
2049
+ "step": 51000
2050
+ },
2051
+ {
2052
+ "epoch": 2.3338016042053744,
2053
+ "grad_norm": 0.8246389031410217,
2054
+ "learning_rate": 1.1103306596577091e-05,
2055
+ "loss": 0.4598,
2056
+ "step": 51500
2057
+ },
2058
+ {
2059
+ "epoch": 2.3338016042053744,
2060
+ "eval_bleu": 47.2042,
2061
+ "eval_chrf++": 66.7914,
2062
+ "eval_gen_len": 27.506,
2063
+ "eval_loss": 0.8135460019111633,
2064
+ "eval_runtime": 727.1036,
2065
+ "eval_samples_per_second": 3.438,
2066
+ "eval_spbleu": 54.8775,
2067
+ "eval_steps_per_second": 0.688,
2068
+ "eval_ter": 43.485,
2069
+ "step": 51500
2070
+ },
2071
+ {
2072
+ "epoch": 2.3564598722073686,
2073
+ "grad_norm": 0.8400627970695496,
2074
+ "learning_rate": 1.072566879654386e-05,
2075
+ "loss": 0.4729,
2076
+ "step": 52000
2077
+ },
2078
+ {
2079
+ "epoch": 2.3564598722073686,
2080
+ "eval_bleu": 45.7548,
2081
+ "eval_chrf++": 66.7646,
2082
+ "eval_gen_len": 27.63,
2083
+ "eval_loss": 0.8120532035827637,
2084
+ "eval_runtime": 730.4307,
2085
+ "eval_samples_per_second": 3.423,
2086
+ "eval_spbleu": 54.8446,
2087
+ "eval_steps_per_second": 0.685,
2088
+ "eval_ter": 43.4956,
2089
+ "step": 52000
2090
+ },
2091
+ {
2092
+ "epoch": 2.3791181402093624,
2093
+ "grad_norm": 0.8089118599891663,
2094
+ "learning_rate": 1.0348030996510627e-05,
2095
+ "loss": 0.4683,
2096
+ "step": 52500
2097
+ },
2098
+ {
2099
+ "epoch": 2.3791181402093624,
2100
+ "eval_bleu": 46.2723,
2101
+ "eval_chrf++": 66.8032,
2102
+ "eval_gen_len": 27.588,
2103
+ "eval_loss": 0.8125736117362976,
2104
+ "eval_runtime": 729.5548,
2105
+ "eval_samples_per_second": 3.427,
2106
+ "eval_spbleu": 54.9204,
2107
+ "eval_steps_per_second": 0.685,
2108
+ "eval_ter": 43.3954,
2109
+ "step": 52500
2110
+ },
2111
+ {
2112
+ "epoch": 2.401776408211356,
2113
+ "grad_norm": 0.6534927487373352,
2114
+ "learning_rate": 9.970393196477396e-06,
2115
+ "loss": 0.4727,
2116
+ "step": 53000
2117
+ },
2118
+ {
2119
+ "epoch": 2.401776408211356,
2120
+ "eval_bleu": 46.7508,
2121
+ "eval_chrf++": 66.7655,
2122
+ "eval_gen_len": 27.5272,
2123
+ "eval_loss": 0.8069682717323303,
2124
+ "eval_runtime": 732.8023,
2125
+ "eval_samples_per_second": 3.412,
2126
+ "eval_spbleu": 54.8765,
2127
+ "eval_steps_per_second": 0.682,
2128
+ "eval_ter": 43.3875,
2129
+ "step": 53000
2130
+ },
2131
+ {
2132
+ "epoch": 2.4244346762133504,
2133
+ "grad_norm": 0.6930407881736755,
2134
+ "learning_rate": 9.592755396444163e-06,
2135
+ "loss": 0.4723,
2136
+ "step": 53500
2137
+ },
2138
+ {
2139
+ "epoch": 2.4244346762133504,
2140
+ "eval_bleu": 47.2069,
2141
+ "eval_chrf++": 66.9501,
2142
+ "eval_gen_len": 27.5308,
2143
+ "eval_loss": 0.8114036321640015,
2144
+ "eval_runtime": 730.3451,
2145
+ "eval_samples_per_second": 3.423,
2146
+ "eval_spbleu": 55.1585,
2147
+ "eval_steps_per_second": 0.685,
2148
+ "eval_ter": 43.2451,
2149
+ "step": 53500
2150
+ },
2151
+ {
2152
+ "epoch": 2.447092944215344,
2153
+ "grad_norm": 0.7665801644325256,
2154
+ "learning_rate": 9.215117596410932e-06,
2155
+ "loss": 0.4711,
2156
+ "step": 54000
2157
+ },
2158
+ {
2159
+ "epoch": 2.447092944215344,
2160
+ "eval_bleu": 47.4976,
2161
+ "eval_chrf++": 67.0709,
2162
+ "eval_gen_len": 27.4768,
2163
+ "eval_loss": 0.8121780753135681,
2164
+ "eval_runtime": 727.4198,
2165
+ "eval_samples_per_second": 3.437,
2166
+ "eval_spbleu": 55.3017,
2167
+ "eval_steps_per_second": 0.687,
2168
+ "eval_ter": 43.0816,
2169
+ "step": 54000
2170
+ }
2171
+ ],
2172
+ "logging_steps": 500,
2173
+ "max_steps": 66201,
2174
+ "num_input_tokens_seen": 0,
2175
+ "num_train_epochs": 3,
2176
+ "save_steps": 200,
2177
+ "stateful_callbacks": {
2178
+ "TrainerControl": {
2179
+ "args": {
2180
+ "should_epoch_stop": false,
2181
+ "should_evaluate": false,
2182
+ "should_log": false,
2183
+ "should_save": true,
2184
+ "should_training_stop": false
2185
+ },
2186
+ "attributes": {}
2187
+ }
2188
+ },
2189
+ "total_flos": 3.756455083596841e+18,
2190
+ "train_batch_size": 32,
2191
+ "trial_name": null,
2192
+ "trial_params": null
2193
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba8601ac68fb7106bc81b74e9e28b1ff81dcf75e6101c57f306eb44c2d9e81fd
3
+ size 5304