chipewyan
commited on
Commit
•
291d340
1
Parent(s):
084795c
-m
Browse files- config.json +116 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +12 -0
- trainer_state.json +1973 -0
- vocab.json +72 -0
config.json
ADDED
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"adapter_kernel_size": 3,
|
5 |
+
"adapter_stride": 2,
|
6 |
+
"add_adapter": false,
|
7 |
+
"apply_spec_augment": true,
|
8 |
+
"architectures": [
|
9 |
+
"Wav2Vec2ForCTC"
|
10 |
+
],
|
11 |
+
"attention_dropout": 0.1,
|
12 |
+
"bos_token_id": 1,
|
13 |
+
"classifier_proj_size": 256,
|
14 |
+
"codevector_dim": 768,
|
15 |
+
"contrastive_logits_temperature": 0.1,
|
16 |
+
"conv_bias": true,
|
17 |
+
"conv_dim": [
|
18 |
+
512,
|
19 |
+
512,
|
20 |
+
512,
|
21 |
+
512,
|
22 |
+
512,
|
23 |
+
512,
|
24 |
+
512
|
25 |
+
],
|
26 |
+
"conv_kernel": [
|
27 |
+
10,
|
28 |
+
3,
|
29 |
+
3,
|
30 |
+
3,
|
31 |
+
3,
|
32 |
+
2,
|
33 |
+
2
|
34 |
+
],
|
35 |
+
"conv_stride": [
|
36 |
+
5,
|
37 |
+
2,
|
38 |
+
2,
|
39 |
+
2,
|
40 |
+
2,
|
41 |
+
2,
|
42 |
+
2
|
43 |
+
],
|
44 |
+
"ctc_loss_reduction": "mean",
|
45 |
+
"ctc_zero_infinity": false,
|
46 |
+
"diversity_loss_weight": 0.1,
|
47 |
+
"do_stable_layer_norm": true,
|
48 |
+
"eos_token_id": 2,
|
49 |
+
"feat_extract_activation": "gelu",
|
50 |
+
"feat_extract_dropout": 0.0,
|
51 |
+
"feat_extract_norm": "layer",
|
52 |
+
"feat_proj_dropout": 0.0,
|
53 |
+
"feat_quantizer_dropout": 0.0,
|
54 |
+
"final_dropout": 0.0,
|
55 |
+
"gradient_checkpointing": false,
|
56 |
+
"hidden_act": "gelu",
|
57 |
+
"hidden_dropout": 0.1,
|
58 |
+
"hidden_size": 1024,
|
59 |
+
"initializer_range": 0.02,
|
60 |
+
"intermediate_size": 4096,
|
61 |
+
"layer_norm_eps": 1e-05,
|
62 |
+
"layerdrop": 0.1,
|
63 |
+
"mask_channel_length": 10,
|
64 |
+
"mask_channel_min_space": 1,
|
65 |
+
"mask_channel_other": 0.0,
|
66 |
+
"mask_channel_prob": 0.0,
|
67 |
+
"mask_channel_selection": "static",
|
68 |
+
"mask_feature_length": 10,
|
69 |
+
"mask_feature_min_masks": 0,
|
70 |
+
"mask_feature_prob": 0.0,
|
71 |
+
"mask_time_length": 10,
|
72 |
+
"mask_time_min_masks": 2,
|
73 |
+
"mask_time_min_space": 1,
|
74 |
+
"mask_time_other": 0.0,
|
75 |
+
"mask_time_prob": 0.05,
|
76 |
+
"mask_time_selection": "static",
|
77 |
+
"model_type": "wav2vec2",
|
78 |
+
"num_adapter_layers": 3,
|
79 |
+
"num_attention_heads": 16,
|
80 |
+
"num_codevector_groups": 2,
|
81 |
+
"num_codevectors_per_group": 320,
|
82 |
+
"num_conv_pos_embedding_groups": 16,
|
83 |
+
"num_conv_pos_embeddings": 128,
|
84 |
+
"num_feat_extract_layers": 7,
|
85 |
+
"num_hidden_layers": 24,
|
86 |
+
"num_negatives": 100,
|
87 |
+
"output_hidden_size": 1024,
|
88 |
+
"pad_token_id": 69,
|
89 |
+
"proj_codevector_dim": 768,
|
90 |
+
"tdnn_dilation": [
|
91 |
+
1,
|
92 |
+
2,
|
93 |
+
3,
|
94 |
+
1,
|
95 |
+
1
|
96 |
+
],
|
97 |
+
"tdnn_dim": [
|
98 |
+
512,
|
99 |
+
512,
|
100 |
+
512,
|
101 |
+
512,
|
102 |
+
1500
|
103 |
+
],
|
104 |
+
"tdnn_kernel": [
|
105 |
+
5,
|
106 |
+
3,
|
107 |
+
3,
|
108 |
+
1,
|
109 |
+
1
|
110 |
+
],
|
111 |
+
"torch_dtype": "float32",
|
112 |
+
"transformers_version": "4.28.1",
|
113 |
+
"use_weighted_layer_sum": false,
|
114 |
+
"vocab_size": 70,
|
115 |
+
"xvector_output_dim": 512
|
116 |
+
}
|
preprocessor_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
4 |
+
"feature_size": 1,
|
5 |
+
"padding_side": "right",
|
6 |
+
"padding_value": 0.0,
|
7 |
+
"return_attention_mask": true,
|
8 |
+
"sampling_rate": 16000
|
9 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b2653dbd4017188e6a771cd5ff797e2a9c448ab1609b132f4344a6f22ce5fd61
|
3 |
+
size 1262182957
|
special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"eos_token": "</s>",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"unk_token": "[UNK]"
|
6 |
+
}
|
tokenizer_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"clean_up_tokenization_spaces": true,
|
4 |
+
"do_lower_case": false,
|
5 |
+
"eos_token": "</s>",
|
6 |
+
"model_max_length": 1000000000000000019884624838656,
|
7 |
+
"pad_token": "[PAD]",
|
8 |
+
"replace_word_delimiter_char": " ",
|
9 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
10 |
+
"unk_token": "[UNK]",
|
11 |
+
"word_delimiter_token": "|"
|
12 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1973 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 9.99825479930192,
|
5 |
+
"global_step": 2860,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 2.9999999999999997e-05,
|
13 |
+
"loss": 12.9083,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.07,
|
18 |
+
"learning_rate": 5.9999999999999995e-05,
|
19 |
+
"loss": 12.5233,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.1,
|
24 |
+
"learning_rate": 8.699999999999999e-05,
|
25 |
+
"loss": 12.1986,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.14,
|
30 |
+
"learning_rate": 0.00011099999999999999,
|
31 |
+
"loss": 9.567,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.17,
|
36 |
+
"learning_rate": 0.000138,
|
37 |
+
"loss": 6.7421,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.21,
|
42 |
+
"learning_rate": 0.000168,
|
43 |
+
"loss": 4.3567,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.24,
|
48 |
+
"learning_rate": 0.000198,
|
49 |
+
"loss": 3.5698,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.28,
|
54 |
+
"learning_rate": 0.00022799999999999999,
|
55 |
+
"loss": 3.2567,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.31,
|
60 |
+
"learning_rate": 0.000258,
|
61 |
+
"loss": 3.145,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.35,
|
66 |
+
"learning_rate": 0.00028799999999999995,
|
67 |
+
"loss": 3.2802,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.35,
|
72 |
+
"eval_loss": 3.8881919384002686,
|
73 |
+
"eval_runtime": 112.7654,
|
74 |
+
"eval_samples_per_second": 12.867,
|
75 |
+
"eval_steps_per_second": 0.807,
|
76 |
+
"step": 100
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.38,
|
80 |
+
"learning_rate": 0.0002993478260869565,
|
81 |
+
"loss": 3.6105,
|
82 |
+
"step": 110
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 0.42,
|
86 |
+
"learning_rate": 0.00029826086956521737,
|
87 |
+
"loss": 3.1704,
|
88 |
+
"step": 120
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.45,
|
92 |
+
"learning_rate": 0.00029717391304347825,
|
93 |
+
"loss": 3.1124,
|
94 |
+
"step": 130
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.49,
|
98 |
+
"learning_rate": 0.0002960869565217391,
|
99 |
+
"loss": 3.1266,
|
100 |
+
"step": 140
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.52,
|
104 |
+
"learning_rate": 0.00029499999999999996,
|
105 |
+
"loss": 3.2521,
|
106 |
+
"step": 150
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.56,
|
110 |
+
"learning_rate": 0.00029391304347826084,
|
111 |
+
"loss": 3.6024,
|
112 |
+
"step": 160
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.59,
|
116 |
+
"learning_rate": 0.0002928260869565217,
|
117 |
+
"loss": 3.2086,
|
118 |
+
"step": 170
|
119 |
+
},
|
120 |
+
{
|
121 |
+
"epoch": 0.63,
|
122 |
+
"learning_rate": 0.0002917391304347826,
|
123 |
+
"loss": 3.1616,
|
124 |
+
"step": 180
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 0.66,
|
128 |
+
"learning_rate": 0.00029065217391304344,
|
129 |
+
"loss": 3.1408,
|
130 |
+
"step": 190
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 0.7,
|
134 |
+
"learning_rate": 0.0002895652173913043,
|
135 |
+
"loss": 3.2254,
|
136 |
+
"step": 200
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.7,
|
140 |
+
"eval_loss": 3.885490894317627,
|
141 |
+
"eval_runtime": 111.0614,
|
142 |
+
"eval_samples_per_second": 13.065,
|
143 |
+
"eval_steps_per_second": 0.819,
|
144 |
+
"step": 200
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.73,
|
148 |
+
"learning_rate": 0.0002884782608695652,
|
149 |
+
"loss": 3.6835,
|
150 |
+
"step": 210
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.77,
|
154 |
+
"learning_rate": 0.00028739130434782603,
|
155 |
+
"loss": 3.1449,
|
156 |
+
"step": 220
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.8,
|
160 |
+
"learning_rate": 0.0002863043478260869,
|
161 |
+
"loss": 3.2003,
|
162 |
+
"step": 230
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.84,
|
166 |
+
"learning_rate": 0.0002852173913043478,
|
167 |
+
"loss": 3.0836,
|
168 |
+
"step": 240
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.87,
|
172 |
+
"learning_rate": 0.0002841304347826087,
|
173 |
+
"loss": 3.1128,
|
174 |
+
"step": 250
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.91,
|
178 |
+
"learning_rate": 0.0002830434782608695,
|
179 |
+
"loss": 3.3853,
|
180 |
+
"step": 260
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 0.94,
|
184 |
+
"learning_rate": 0.0002819565217391304,
|
185 |
+
"loss": 3.102,
|
186 |
+
"step": 270
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 0.98,
|
190 |
+
"learning_rate": 0.0002808695652173913,
|
191 |
+
"loss": 3.0516,
|
192 |
+
"step": 280
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 1.01,
|
196 |
+
"learning_rate": 0.0002797826086956521,
|
197 |
+
"loss": 3.5681,
|
198 |
+
"step": 290
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 1.05,
|
202 |
+
"learning_rate": 0.00027869565217391305,
|
203 |
+
"loss": 3.1697,
|
204 |
+
"step": 300
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 1.05,
|
208 |
+
"eval_loss": 3.419605255126953,
|
209 |
+
"eval_runtime": 111.0117,
|
210 |
+
"eval_samples_per_second": 13.071,
|
211 |
+
"eval_steps_per_second": 0.82,
|
212 |
+
"step": 300
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 1.08,
|
216 |
+
"learning_rate": 0.0002776086956521739,
|
217 |
+
"loss": 3.0489,
|
218 |
+
"step": 310
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 1.12,
|
222 |
+
"learning_rate": 0.00027652173913043476,
|
223 |
+
"loss": 3.0347,
|
224 |
+
"step": 320
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 1.15,
|
228 |
+
"learning_rate": 0.00027543478260869564,
|
229 |
+
"loss": 3.0201,
|
230 |
+
"step": 330
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 1.19,
|
234 |
+
"learning_rate": 0.00027434782608695647,
|
235 |
+
"loss": 3.2079,
|
236 |
+
"step": 340
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 1.22,
|
240 |
+
"learning_rate": 0.00027326086956521735,
|
241 |
+
"loss": 3.0785,
|
242 |
+
"step": 350
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 1.26,
|
246 |
+
"learning_rate": 0.00027217391304347824,
|
247 |
+
"loss": 2.9747,
|
248 |
+
"step": 360
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 1.29,
|
252 |
+
"learning_rate": 0.0002710869565217391,
|
253 |
+
"loss": 2.9664,
|
254 |
+
"step": 370
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 1.33,
|
258 |
+
"learning_rate": 0.00027,
|
259 |
+
"loss": 2.9724,
|
260 |
+
"step": 380
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 1.36,
|
264 |
+
"learning_rate": 0.00026891304347826083,
|
265 |
+
"loss": 3.1082,
|
266 |
+
"step": 390
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 1.4,
|
270 |
+
"learning_rate": 0.0002678260869565217,
|
271 |
+
"loss": 3.0258,
|
272 |
+
"step": 400
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 1.4,
|
276 |
+
"eval_loss": 3.195385217666626,
|
277 |
+
"eval_runtime": 110.895,
|
278 |
+
"eval_samples_per_second": 13.084,
|
279 |
+
"eval_steps_per_second": 0.821,
|
280 |
+
"step": 400
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 1.43,
|
284 |
+
"learning_rate": 0.0002667391304347826,
|
285 |
+
"loss": 2.9719,
|
286 |
+
"step": 410
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 1.47,
|
290 |
+
"learning_rate": 0.00026565217391304343,
|
291 |
+
"loss": 2.9884,
|
292 |
+
"step": 420
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 1.5,
|
296 |
+
"learning_rate": 0.00026456521739130437,
|
297 |
+
"loss": 2.9625,
|
298 |
+
"step": 430
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 1.54,
|
302 |
+
"learning_rate": 0.0002634782608695652,
|
303 |
+
"loss": 3.0735,
|
304 |
+
"step": 440
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.57,
|
308 |
+
"learning_rate": 0.0002623913043478261,
|
309 |
+
"loss": 2.9998,
|
310 |
+
"step": 450
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 1.61,
|
314 |
+
"learning_rate": 0.00026130434782608696,
|
315 |
+
"loss": 2.9621,
|
316 |
+
"step": 460
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.64,
|
320 |
+
"learning_rate": 0.0002602173913043478,
|
321 |
+
"loss": 2.9433,
|
322 |
+
"step": 470
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 1.68,
|
326 |
+
"learning_rate": 0.0002591304347826087,
|
327 |
+
"loss": 2.9256,
|
328 |
+
"step": 480
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 1.71,
|
332 |
+
"learning_rate": 0.00025804347826086956,
|
333 |
+
"loss": 2.9996,
|
334 |
+
"step": 490
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"epoch": 1.75,
|
338 |
+
"learning_rate": 0.00025695652173913044,
|
339 |
+
"loss": 2.8553,
|
340 |
+
"step": 500
|
341 |
+
},
|
342 |
+
{
|
343 |
+
"epoch": 1.75,
|
344 |
+
"eval_loss": 3.017104387283325,
|
345 |
+
"eval_runtime": 110.9455,
|
346 |
+
"eval_samples_per_second": 13.078,
|
347 |
+
"eval_steps_per_second": 0.82,
|
348 |
+
"step": 500
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 1.78,
|
352 |
+
"learning_rate": 0.00025586956521739127,
|
353 |
+
"loss": 2.7029,
|
354 |
+
"step": 510
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 1.82,
|
358 |
+
"learning_rate": 0.00025478260869565215,
|
359 |
+
"loss": 2.4464,
|
360 |
+
"step": 520
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.85,
|
364 |
+
"learning_rate": 0.00025369565217391303,
|
365 |
+
"loss": 2.2426,
|
366 |
+
"step": 530
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 1.89,
|
370 |
+
"learning_rate": 0.00025260869565217386,
|
371 |
+
"loss": 2.2914,
|
372 |
+
"step": 540
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 1.92,
|
376 |
+
"learning_rate": 0.00025152173913043475,
|
377 |
+
"loss": 1.9541,
|
378 |
+
"step": 550
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 1.96,
|
382 |
+
"learning_rate": 0.00025043478260869563,
|
383 |
+
"loss": 1.5822,
|
384 |
+
"step": 560
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 1.99,
|
388 |
+
"learning_rate": 0.0002493478260869565,
|
389 |
+
"loss": 1.5446,
|
390 |
+
"step": 570
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 2.03,
|
394 |
+
"learning_rate": 0.0002482608695652174,
|
395 |
+
"loss": 2.0892,
|
396 |
+
"step": 580
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 2.06,
|
400 |
+
"learning_rate": 0.0002471739130434782,
|
401 |
+
"loss": 1.3977,
|
402 |
+
"step": 590
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 2.1,
|
406 |
+
"learning_rate": 0.0002460869565217391,
|
407 |
+
"loss": 1.2436,
|
408 |
+
"step": 600
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 2.1,
|
412 |
+
"eval_loss": 2.155662775039673,
|
413 |
+
"eval_runtime": 111.1811,
|
414 |
+
"eval_samples_per_second": 13.051,
|
415 |
+
"eval_steps_per_second": 0.818,
|
416 |
+
"step": 600
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 2.13,
|
420 |
+
"learning_rate": 0.000245,
|
421 |
+
"loss": 1.1344,
|
422 |
+
"step": 610
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 2.17,
|
426 |
+
"learning_rate": 0.00024391304347826085,
|
427 |
+
"loss": 1.1795,
|
428 |
+
"step": 620
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 2.2,
|
432 |
+
"learning_rate": 0.0002428260869565217,
|
433 |
+
"loss": 2.1046,
|
434 |
+
"step": 630
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 2.24,
|
438 |
+
"learning_rate": 0.00024173913043478261,
|
439 |
+
"loss": 1.0792,
|
440 |
+
"step": 640
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 2.27,
|
444 |
+
"learning_rate": 0.00024065217391304347,
|
445 |
+
"loss": 1.157,
|
446 |
+
"step": 650
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 2.31,
|
450 |
+
"learning_rate": 0.00023956521739130433,
|
451 |
+
"loss": 1.0428,
|
452 |
+
"step": 660
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 2.34,
|
456 |
+
"learning_rate": 0.00023847826086956518,
|
457 |
+
"loss": 1.1063,
|
458 |
+
"step": 670
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 2.38,
|
462 |
+
"learning_rate": 0.00023739130434782607,
|
463 |
+
"loss": 1.6716,
|
464 |
+
"step": 680
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 2.41,
|
468 |
+
"learning_rate": 0.00023630434782608692,
|
469 |
+
"loss": 0.9602,
|
470 |
+
"step": 690
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 2.45,
|
474 |
+
"learning_rate": 0.00023521739130434778,
|
475 |
+
"loss": 0.9642,
|
476 |
+
"step": 700
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 2.45,
|
480 |
+
"eval_loss": 1.5505317449569702,
|
481 |
+
"eval_runtime": 111.9226,
|
482 |
+
"eval_samples_per_second": 12.964,
|
483 |
+
"eval_steps_per_second": 0.813,
|
484 |
+
"step": 700
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 2.48,
|
488 |
+
"learning_rate": 0.0002341304347826087,
|
489 |
+
"loss": 0.8874,
|
490 |
+
"step": 710
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 2.52,
|
494 |
+
"learning_rate": 0.00023304347826086954,
|
495 |
+
"loss": 0.9596,
|
496 |
+
"step": 720
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 2.55,
|
500 |
+
"learning_rate": 0.00023195652173913043,
|
501 |
+
"loss": 1.3932,
|
502 |
+
"step": 730
|
503 |
+
},
|
504 |
+
{
|
505 |
+
"epoch": 2.59,
|
506 |
+
"learning_rate": 0.00023086956521739128,
|
507 |
+
"loss": 0.8699,
|
508 |
+
"step": 740
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 2.62,
|
512 |
+
"learning_rate": 0.00022978260869565214,
|
513 |
+
"loss": 0.8153,
|
514 |
+
"step": 750
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 2.66,
|
518 |
+
"learning_rate": 0.00022869565217391302,
|
519 |
+
"loss": 0.9178,
|
520 |
+
"step": 760
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 2.69,
|
524 |
+
"learning_rate": 0.00022760869565217388,
|
525 |
+
"loss": 0.9647,
|
526 |
+
"step": 770
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 2.73,
|
530 |
+
"learning_rate": 0.00022652173913043476,
|
531 |
+
"loss": 1.0555,
|
532 |
+
"step": 780
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 2.76,
|
536 |
+
"learning_rate": 0.00022543478260869565,
|
537 |
+
"loss": 0.8834,
|
538 |
+
"step": 790
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"epoch": 2.8,
|
542 |
+
"learning_rate": 0.0002243478260869565,
|
543 |
+
"loss": 0.752,
|
544 |
+
"step": 800
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 2.8,
|
548 |
+
"eval_loss": 0.9008368253707886,
|
549 |
+
"eval_runtime": 111.0482,
|
550 |
+
"eval_samples_per_second": 13.066,
|
551 |
+
"eval_steps_per_second": 0.819,
|
552 |
+
"step": 800
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 2.83,
|
556 |
+
"learning_rate": 0.00022326086956521736,
|
557 |
+
"loss": 0.8197,
|
558 |
+
"step": 810
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 2.87,
|
562 |
+
"learning_rate": 0.00022217391304347824,
|
563 |
+
"loss": 0.7771,
|
564 |
+
"step": 820
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 2.9,
|
568 |
+
"learning_rate": 0.0002210869565217391,
|
569 |
+
"loss": 0.9506,
|
570 |
+
"step": 830
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 2.94,
|
574 |
+
"learning_rate": 0.00021999999999999995,
|
575 |
+
"loss": 0.7518,
|
576 |
+
"step": 840
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 2.97,
|
580 |
+
"learning_rate": 0.00021891304347826086,
|
581 |
+
"loss": 0.7297,
|
582 |
+
"step": 850
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 3.01,
|
586 |
+
"learning_rate": 0.00021782608695652172,
|
587 |
+
"loss": 0.8188,
|
588 |
+
"step": 860
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 3.04,
|
592 |
+
"learning_rate": 0.0002167391304347826,
|
593 |
+
"loss": 0.6478,
|
594 |
+
"step": 870
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 3.08,
|
598 |
+
"learning_rate": 0.00021565217391304346,
|
599 |
+
"loss": 0.6546,
|
600 |
+
"step": 880
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 3.11,
|
604 |
+
"learning_rate": 0.00021456521739130432,
|
605 |
+
"loss": 0.7012,
|
606 |
+
"step": 890
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 3.15,
|
610 |
+
"learning_rate": 0.0002134782608695652,
|
611 |
+
"loss": 0.6164,
|
612 |
+
"step": 900
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 3.15,
|
616 |
+
"eval_loss": 0.6567979454994202,
|
617 |
+
"eval_runtime": 111.2282,
|
618 |
+
"eval_samples_per_second": 13.045,
|
619 |
+
"eval_steps_per_second": 0.818,
|
620 |
+
"step": 900
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 3.18,
|
624 |
+
"learning_rate": 0.00021239130434782605,
|
625 |
+
"loss": 0.7296,
|
626 |
+
"step": 910
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 3.22,
|
630 |
+
"learning_rate": 0.00021130434782608694,
|
631 |
+
"loss": 0.6166,
|
632 |
+
"step": 920
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 3.25,
|
636 |
+
"learning_rate": 0.00021021739130434782,
|
637 |
+
"loss": 0.6354,
|
638 |
+
"step": 930
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 3.29,
|
642 |
+
"learning_rate": 0.00020913043478260868,
|
643 |
+
"loss": 0.6388,
|
644 |
+
"step": 940
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 3.32,
|
648 |
+
"learning_rate": 0.00020804347826086953,
|
649 |
+
"loss": 0.6069,
|
650 |
+
"step": 950
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 3.36,
|
654 |
+
"learning_rate": 0.00020695652173913042,
|
655 |
+
"loss": 0.7107,
|
656 |
+
"step": 960
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 3.39,
|
660 |
+
"learning_rate": 0.00020586956521739127,
|
661 |
+
"loss": 0.6162,
|
662 |
+
"step": 970
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 3.43,
|
666 |
+
"learning_rate": 0.00020478260869565213,
|
667 |
+
"loss": 0.567,
|
668 |
+
"step": 980
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 3.46,
|
672 |
+
"learning_rate": 0.00020369565217391304,
|
673 |
+
"loss": 0.5679,
|
674 |
+
"step": 990
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 3.5,
|
678 |
+
"learning_rate": 0.0002026086956521739,
|
679 |
+
"loss": 0.6362,
|
680 |
+
"step": 1000
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 3.5,
|
684 |
+
"eval_loss": 0.5460886359214783,
|
685 |
+
"eval_runtime": 110.911,
|
686 |
+
"eval_samples_per_second": 13.083,
|
687 |
+
"eval_steps_per_second": 0.82,
|
688 |
+
"step": 1000
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 3.53,
|
692 |
+
"learning_rate": 0.00020152173913043478,
|
693 |
+
"loss": 0.6771,
|
694 |
+
"step": 1010
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 3.57,
|
698 |
+
"learning_rate": 0.00020043478260869563,
|
699 |
+
"loss": 0.6144,
|
700 |
+
"step": 1020
|
701 |
+
},
|
702 |
+
{
|
703 |
+
"epoch": 3.6,
|
704 |
+
"learning_rate": 0.0001993478260869565,
|
705 |
+
"loss": 0.583,
|
706 |
+
"step": 1030
|
707 |
+
},
|
708 |
+
{
|
709 |
+
"epoch": 3.64,
|
710 |
+
"learning_rate": 0.00019826086956521737,
|
711 |
+
"loss": 0.5773,
|
712 |
+
"step": 1040
|
713 |
+
},
|
714 |
+
{
|
715 |
+
"epoch": 3.67,
|
716 |
+
"learning_rate": 0.00019717391304347826,
|
717 |
+
"loss": 0.5981,
|
718 |
+
"step": 1050
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 3.71,
|
722 |
+
"learning_rate": 0.0001960869565217391,
|
723 |
+
"loss": 0.6162,
|
724 |
+
"step": 1060
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 3.74,
|
728 |
+
"learning_rate": 0.000195,
|
729 |
+
"loss": 0.5387,
|
730 |
+
"step": 1070
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 3.77,
|
734 |
+
"learning_rate": 0.00019391304347826085,
|
735 |
+
"loss": 0.5651,
|
736 |
+
"step": 1080
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 3.81,
|
740 |
+
"learning_rate": 0.0001928260869565217,
|
741 |
+
"loss": 0.544,
|
742 |
+
"step": 1090
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 3.84,
|
746 |
+
"learning_rate": 0.0001917391304347826,
|
747 |
+
"loss": 0.5539,
|
748 |
+
"step": 1100
|
749 |
+
},
|
750 |
+
{
|
751 |
+
"epoch": 3.84,
|
752 |
+
"eval_loss": 0.5009284615516663,
|
753 |
+
"eval_runtime": 111.9745,
|
754 |
+
"eval_samples_per_second": 12.958,
|
755 |
+
"eval_steps_per_second": 0.813,
|
756 |
+
"step": 1100
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 3.88,
|
760 |
+
"learning_rate": 0.00019065217391304345,
|
761 |
+
"loss": 0.6894,
|
762 |
+
"step": 1110
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 3.91,
|
766 |
+
"learning_rate": 0.00018956521739130436,
|
767 |
+
"loss": 0.5409,
|
768 |
+
"step": 1120
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 3.95,
|
772 |
+
"learning_rate": 0.00018847826086956521,
|
773 |
+
"loss": 0.5676,
|
774 |
+
"step": 1130
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 3.98,
|
778 |
+
"learning_rate": 0.00018739130434782607,
|
779 |
+
"loss": 0.5543,
|
780 |
+
"step": 1140
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 4.02,
|
784 |
+
"learning_rate": 0.00018630434782608695,
|
785 |
+
"loss": 0.6413,
|
786 |
+
"step": 1150
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 4.06,
|
790 |
+
"learning_rate": 0.0001852173913043478,
|
791 |
+
"loss": 0.4664,
|
792 |
+
"step": 1160
|
793 |
+
},
|
794 |
+
{
|
795 |
+
"epoch": 4.09,
|
796 |
+
"learning_rate": 0.00018413043478260867,
|
797 |
+
"loss": 0.5098,
|
798 |
+
"step": 1170
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 4.13,
|
802 |
+
"learning_rate": 0.00018304347826086955,
|
803 |
+
"loss": 0.4832,
|
804 |
+
"step": 1180
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 4.16,
|
808 |
+
"learning_rate": 0.00018195652173913043,
|
809 |
+
"loss": 0.4859,
|
810 |
+
"step": 1190
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 4.2,
|
814 |
+
"learning_rate": 0.0001808695652173913,
|
815 |
+
"loss": 0.5144,
|
816 |
+
"step": 1200
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 4.2,
|
820 |
+
"eval_loss": 0.4911543130874634,
|
821 |
+
"eval_runtime": 111.7063,
|
822 |
+
"eval_samples_per_second": 12.989,
|
823 |
+
"eval_steps_per_second": 0.815,
|
824 |
+
"step": 1200
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 4.23,
|
828 |
+
"learning_rate": 0.00017978260869565217,
|
829 |
+
"loss": 0.4853,
|
830 |
+
"step": 1210
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 4.27,
|
834 |
+
"learning_rate": 0.00017869565217391303,
|
835 |
+
"loss": 0.5149,
|
836 |
+
"step": 1220
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 4.3,
|
840 |
+
"learning_rate": 0.00017760869565217388,
|
841 |
+
"loss": 0.4608,
|
842 |
+
"step": 1230
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 4.34,
|
846 |
+
"learning_rate": 0.00017652173913043477,
|
847 |
+
"loss": 0.4917,
|
848 |
+
"step": 1240
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 4.37,
|
852 |
+
"learning_rate": 0.00017543478260869562,
|
853 |
+
"loss": 0.4615,
|
854 |
+
"step": 1250
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 4.4,
|
858 |
+
"learning_rate": 0.00017434782608695653,
|
859 |
+
"loss": 0.4175,
|
860 |
+
"step": 1260
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 4.44,
|
864 |
+
"learning_rate": 0.0001732608695652174,
|
865 |
+
"loss": 0.4752,
|
866 |
+
"step": 1270
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 4.47,
|
870 |
+
"learning_rate": 0.00017217391304347825,
|
871 |
+
"loss": 0.4979,
|
872 |
+
"step": 1280
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 4.51,
|
876 |
+
"learning_rate": 0.00017108695652173913,
|
877 |
+
"loss": 0.4272,
|
878 |
+
"step": 1290
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 4.54,
|
882 |
+
"learning_rate": 0.00016999999999999999,
|
883 |
+
"loss": 0.5067,
|
884 |
+
"step": 1300
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 4.54,
|
888 |
+
"eval_loss": 0.4602111279964447,
|
889 |
+
"eval_runtime": 111.5487,
|
890 |
+
"eval_samples_per_second": 13.008,
|
891 |
+
"eval_steps_per_second": 0.816,
|
892 |
+
"step": 1300
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 4.58,
|
896 |
+
"learning_rate": 0.00016891304347826084,
|
897 |
+
"loss": 0.483,
|
898 |
+
"step": 1310
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 4.61,
|
902 |
+
"learning_rate": 0.0001678260869565217,
|
903 |
+
"loss": 0.4671,
|
904 |
+
"step": 1320
|
905 |
+
},
|
906 |
+
{
|
907 |
+
"epoch": 4.65,
|
908 |
+
"learning_rate": 0.0001667391304347826,
|
909 |
+
"loss": 0.4806,
|
910 |
+
"step": 1330
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 4.68,
|
914 |
+
"learning_rate": 0.00016565217391304346,
|
915 |
+
"loss": 0.4798,
|
916 |
+
"step": 1340
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 4.72,
|
920 |
+
"learning_rate": 0.00016456521739130435,
|
921 |
+
"loss": 0.5021,
|
922 |
+
"step": 1350
|
923 |
+
},
|
924 |
+
{
|
925 |
+
"epoch": 4.75,
|
926 |
+
"learning_rate": 0.0001634782608695652,
|
927 |
+
"loss": 0.4445,
|
928 |
+
"step": 1360
|
929 |
+
},
|
930 |
+
{
|
931 |
+
"epoch": 4.79,
|
932 |
+
"learning_rate": 0.00016239130434782606,
|
933 |
+
"loss": 0.4462,
|
934 |
+
"step": 1370
|
935 |
+
},
|
936 |
+
{
|
937 |
+
"epoch": 4.82,
|
938 |
+
"learning_rate": 0.00016130434782608694,
|
939 |
+
"loss": 0.5033,
|
940 |
+
"step": 1380
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 4.86,
|
944 |
+
"learning_rate": 0.0001602173913043478,
|
945 |
+
"loss": 0.4577,
|
946 |
+
"step": 1390
|
947 |
+
},
|
948 |
+
{
|
949 |
+
"epoch": 4.89,
|
950 |
+
"learning_rate": 0.0001591304347826087,
|
951 |
+
"loss": 0.4999,
|
952 |
+
"step": 1400
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 4.89,
|
956 |
+
"eval_loss": 0.4281094968318939,
|
957 |
+
"eval_runtime": 111.6556,
|
958 |
+
"eval_samples_per_second": 12.995,
|
959 |
+
"eval_steps_per_second": 0.815,
|
960 |
+
"step": 1400
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 4.93,
|
964 |
+
"learning_rate": 0.00015804347826086956,
|
965 |
+
"loss": 0.4366,
|
966 |
+
"step": 1410
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 4.96,
|
970 |
+
"learning_rate": 0.00015695652173913042,
|
971 |
+
"loss": 0.4696,
|
972 |
+
"step": 1420
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 5.0,
|
976 |
+
"learning_rate": 0.00015586956521739128,
|
977 |
+
"loss": 0.491,
|
978 |
+
"step": 1430
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 5.03,
|
982 |
+
"learning_rate": 0.00015478260869565216,
|
983 |
+
"loss": 0.4202,
|
984 |
+
"step": 1440
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 5.07,
|
988 |
+
"learning_rate": 0.00015369565217391302,
|
989 |
+
"loss": 0.4307,
|
990 |
+
"step": 1450
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 5.1,
|
994 |
+
"learning_rate": 0.00015260869565217387,
|
995 |
+
"loss": 0.4103,
|
996 |
+
"step": 1460
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 5.14,
|
1000 |
+
"learning_rate": 0.00015152173913043478,
|
1001 |
+
"loss": 0.4157,
|
1002 |
+
"step": 1470
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 5.17,
|
1006 |
+
"learning_rate": 0.00015043478260869564,
|
1007 |
+
"loss": 0.4326,
|
1008 |
+
"step": 1480
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 5.21,
|
1012 |
+
"learning_rate": 0.00014934782608695652,
|
1013 |
+
"loss": 0.3842,
|
1014 |
+
"step": 1490
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 5.24,
|
1018 |
+
"learning_rate": 0.00014826086956521738,
|
1019 |
+
"loss": 0.4072,
|
1020 |
+
"step": 1500
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 5.24,
|
1024 |
+
"eval_loss": 0.416363924741745,
|
1025 |
+
"eval_runtime": 111.6811,
|
1026 |
+
"eval_samples_per_second": 12.992,
|
1027 |
+
"eval_steps_per_second": 0.815,
|
1028 |
+
"step": 1500
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 5.28,
|
1032 |
+
"learning_rate": 0.00014717391304347823,
|
1033 |
+
"loss": 0.4093,
|
1034 |
+
"step": 1510
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 5.31,
|
1038 |
+
"learning_rate": 0.00014608695652173912,
|
1039 |
+
"loss": 0.4013,
|
1040 |
+
"step": 1520
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 5.35,
|
1044 |
+
"learning_rate": 0.000145,
|
1045 |
+
"loss": 0.4534,
|
1046 |
+
"step": 1530
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 5.38,
|
1050 |
+
"learning_rate": 0.00014391304347826086,
|
1051 |
+
"loss": 0.3509,
|
1052 |
+
"step": 1540
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 5.42,
|
1056 |
+
"learning_rate": 0.00014282608695652174,
|
1057 |
+
"loss": 0.4058,
|
1058 |
+
"step": 1550
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 5.45,
|
1062 |
+
"learning_rate": 0.0001417391304347826,
|
1063 |
+
"loss": 0.3956,
|
1064 |
+
"step": 1560
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 5.49,
|
1068 |
+
"learning_rate": 0.00014065217391304345,
|
1069 |
+
"loss": 0.415,
|
1070 |
+
"step": 1570
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 5.52,
|
1074 |
+
"learning_rate": 0.00013956521739130434,
|
1075 |
+
"loss": 0.4255,
|
1076 |
+
"step": 1580
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 5.56,
|
1080 |
+
"learning_rate": 0.00013847826086956522,
|
1081 |
+
"loss": 0.3925,
|
1082 |
+
"step": 1590
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 5.59,
|
1086 |
+
"learning_rate": 0.00013739130434782607,
|
1087 |
+
"loss": 0.3855,
|
1088 |
+
"step": 1600
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 5.59,
|
1092 |
+
"eval_loss": 0.43422192335128784,
|
1093 |
+
"eval_runtime": 111.3707,
|
1094 |
+
"eval_samples_per_second": 13.029,
|
1095 |
+
"eval_steps_per_second": 0.817,
|
1096 |
+
"step": 1600
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 5.63,
|
1100 |
+
"learning_rate": 0.00013630434782608693,
|
1101 |
+
"loss": 0.4364,
|
1102 |
+
"step": 1610
|
1103 |
+
},
|
1104 |
+
{
|
1105 |
+
"epoch": 5.66,
|
1106 |
+
"learning_rate": 0.00013521739130434781,
|
1107 |
+
"loss": 0.3908,
|
1108 |
+
"step": 1620
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 5.7,
|
1112 |
+
"learning_rate": 0.0001341304347826087,
|
1113 |
+
"loss": 0.4419,
|
1114 |
+
"step": 1630
|
1115 |
+
},
|
1116 |
+
{
|
1117 |
+
"epoch": 5.73,
|
1118 |
+
"learning_rate": 0.00013304347826086955,
|
1119 |
+
"loss": 0.3667,
|
1120 |
+
"step": 1640
|
1121 |
+
},
|
1122 |
+
{
|
1123 |
+
"epoch": 5.77,
|
1124 |
+
"learning_rate": 0.0001319565217391304,
|
1125 |
+
"loss": 0.3971,
|
1126 |
+
"step": 1650
|
1127 |
+
},
|
1128 |
+
{
|
1129 |
+
"epoch": 5.8,
|
1130 |
+
"learning_rate": 0.0001308695652173913,
|
1131 |
+
"loss": 0.4102,
|
1132 |
+
"step": 1660
|
1133 |
+
},
|
1134 |
+
{
|
1135 |
+
"epoch": 5.84,
|
1136 |
+
"learning_rate": 0.00012978260869565218,
|
1137 |
+
"loss": 0.4063,
|
1138 |
+
"step": 1670
|
1139 |
+
},
|
1140 |
+
{
|
1141 |
+
"epoch": 5.87,
|
1142 |
+
"learning_rate": 0.00012869565217391303,
|
1143 |
+
"loss": 0.4576,
|
1144 |
+
"step": 1680
|
1145 |
+
},
|
1146 |
+
{
|
1147 |
+
"epoch": 5.91,
|
1148 |
+
"learning_rate": 0.00012760869565217392,
|
1149 |
+
"loss": 0.3524,
|
1150 |
+
"step": 1690
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 5.94,
|
1154 |
+
"learning_rate": 0.00012652173913043477,
|
1155 |
+
"loss": 0.4075,
|
1156 |
+
"step": 1700
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 5.94,
|
1160 |
+
"eval_loss": 0.40860703587532043,
|
1161 |
+
"eval_runtime": 112.0658,
|
1162 |
+
"eval_samples_per_second": 12.948,
|
1163 |
+
"eval_steps_per_second": 0.812,
|
1164 |
+
"step": 1700
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 5.98,
|
1168 |
+
"learning_rate": 0.00012543478260869563,
|
1169 |
+
"loss": 0.3744,
|
1170 |
+
"step": 1710
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 6.01,
|
1174 |
+
"learning_rate": 0.0001243478260869565,
|
1175 |
+
"loss": 0.3863,
|
1176 |
+
"step": 1720
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 6.05,
|
1180 |
+
"learning_rate": 0.0001232608695652174,
|
1181 |
+
"loss": 0.338,
|
1182 |
+
"step": 1730
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 6.08,
|
1186 |
+
"learning_rate": 0.00012217391304347825,
|
1187 |
+
"loss": 0.3789,
|
1188 |
+
"step": 1740
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 6.12,
|
1192 |
+
"learning_rate": 0.00012108695652173912,
|
1193 |
+
"loss": 0.3253,
|
1194 |
+
"step": 1750
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 6.15,
|
1198 |
+
"learning_rate": 0.00011999999999999999,
|
1199 |
+
"loss": 0.3878,
|
1200 |
+
"step": 1760
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 6.19,
|
1204 |
+
"learning_rate": 0.00011891304347826086,
|
1205 |
+
"loss": 0.3732,
|
1206 |
+
"step": 1770
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 6.22,
|
1210 |
+
"learning_rate": 0.00011782608695652173,
|
1211 |
+
"loss": 0.3494,
|
1212 |
+
"step": 1780
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 6.26,
|
1216 |
+
"learning_rate": 0.00011673913043478258,
|
1217 |
+
"loss": 0.3701,
|
1218 |
+
"step": 1790
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 6.29,
|
1222 |
+
"learning_rate": 0.00011565217391304347,
|
1223 |
+
"loss": 0.347,
|
1224 |
+
"step": 1800
|
1225 |
+
},
|
1226 |
+
{
|
1227 |
+
"epoch": 6.29,
|
1228 |
+
"eval_loss": 0.42197009921073914,
|
1229 |
+
"eval_runtime": 111.9078,
|
1230 |
+
"eval_samples_per_second": 12.966,
|
1231 |
+
"eval_steps_per_second": 0.813,
|
1232 |
+
"step": 1800
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 6.33,
|
1236 |
+
"learning_rate": 0.00011456521739130434,
|
1237 |
+
"loss": 0.3484,
|
1238 |
+
"step": 1810
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 6.36,
|
1242 |
+
"learning_rate": 0.00011347826086956521,
|
1243 |
+
"loss": 0.3328,
|
1244 |
+
"step": 1820
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 6.4,
|
1248 |
+
"learning_rate": 0.00011239130434782608,
|
1249 |
+
"loss": 0.3444,
|
1250 |
+
"step": 1830
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 6.43,
|
1254 |
+
"learning_rate": 0.00011130434782608695,
|
1255 |
+
"loss": 0.321,
|
1256 |
+
"step": 1840
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 6.47,
|
1260 |
+
"learning_rate": 0.00011021739130434782,
|
1261 |
+
"loss": 0.4089,
|
1262 |
+
"step": 1850
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 6.5,
|
1266 |
+
"learning_rate": 0.00010913043478260867,
|
1267 |
+
"loss": 0.335,
|
1268 |
+
"step": 1860
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 6.54,
|
1272 |
+
"learning_rate": 0.00010804347826086956,
|
1273 |
+
"loss": 0.376,
|
1274 |
+
"step": 1870
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 6.57,
|
1278 |
+
"learning_rate": 0.00010695652173913043,
|
1279 |
+
"loss": 0.3324,
|
1280 |
+
"step": 1880
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 6.61,
|
1284 |
+
"learning_rate": 0.0001058695652173913,
|
1285 |
+
"loss": 0.3359,
|
1286 |
+
"step": 1890
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 6.64,
|
1290 |
+
"learning_rate": 0.00010478260869565216,
|
1291 |
+
"loss": 0.3708,
|
1292 |
+
"step": 1900
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 6.64,
|
1296 |
+
"eval_loss": 0.4211980998516083,
|
1297 |
+
"eval_runtime": 113.1099,
|
1298 |
+
"eval_samples_per_second": 12.828,
|
1299 |
+
"eval_steps_per_second": 0.805,
|
1300 |
+
"step": 1900
|
1301 |
+
},
|
1302 |
+
{
|
1303 |
+
"epoch": 6.68,
|
1304 |
+
"learning_rate": 0.00010369565217391303,
|
1305 |
+
"loss": 0.3512,
|
1306 |
+
"step": 1910
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 6.71,
|
1310 |
+
"learning_rate": 0.0001026086956521739,
|
1311 |
+
"loss": 0.3381,
|
1312 |
+
"step": 1920
|
1313 |
+
},
|
1314 |
+
{
|
1315 |
+
"epoch": 6.75,
|
1316 |
+
"learning_rate": 0.00010152173913043479,
|
1317 |
+
"loss": 0.2925,
|
1318 |
+
"step": 1930
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 6.78,
|
1322 |
+
"learning_rate": 0.00010043478260869564,
|
1323 |
+
"loss": 0.3805,
|
1324 |
+
"step": 1940
|
1325 |
+
},
|
1326 |
+
{
|
1327 |
+
"epoch": 6.82,
|
1328 |
+
"learning_rate": 9.934782608695651e-05,
|
1329 |
+
"loss": 0.3803,
|
1330 |
+
"step": 1950
|
1331 |
+
},
|
1332 |
+
{
|
1333 |
+
"epoch": 6.85,
|
1334 |
+
"learning_rate": 9.826086956521738e-05,
|
1335 |
+
"loss": 0.3499,
|
1336 |
+
"step": 1960
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 6.89,
|
1340 |
+
"learning_rate": 9.717391304347825e-05,
|
1341 |
+
"loss": 0.3419,
|
1342 |
+
"step": 1970
|
1343 |
+
},
|
1344 |
+
{
|
1345 |
+
"epoch": 6.92,
|
1346 |
+
"learning_rate": 9.608695652173912e-05,
|
1347 |
+
"loss": 0.2918,
|
1348 |
+
"step": 1980
|
1349 |
+
},
|
1350 |
+
{
|
1351 |
+
"epoch": 6.96,
|
1352 |
+
"learning_rate": 9.499999999999999e-05,
|
1353 |
+
"loss": 0.3602,
|
1354 |
+
"step": 1990
|
1355 |
+
},
|
1356 |
+
{
|
1357 |
+
"epoch": 6.99,
|
1358 |
+
"learning_rate": 9.391304347826087e-05,
|
1359 |
+
"loss": 0.3788,
|
1360 |
+
"step": 2000
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 6.99,
|
1364 |
+
"eval_loss": 0.419994056224823,
|
1365 |
+
"eval_runtime": 111.5298,
|
1366 |
+
"eval_samples_per_second": 13.01,
|
1367 |
+
"eval_steps_per_second": 0.816,
|
1368 |
+
"step": 2000
|
1369 |
+
},
|
1370 |
+
{
|
1371 |
+
"epoch": 7.03,
|
1372 |
+
"learning_rate": 9.282608695652173e-05,
|
1373 |
+
"loss": 0.3368,
|
1374 |
+
"step": 2010
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 7.06,
|
1378 |
+
"learning_rate": 9.17391304347826e-05,
|
1379 |
+
"loss": 0.3186,
|
1380 |
+
"step": 2020
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 7.1,
|
1384 |
+
"learning_rate": 9.065217391304346e-05,
|
1385 |
+
"loss": 0.3134,
|
1386 |
+
"step": 2030
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 7.13,
|
1390 |
+
"learning_rate": 8.956521739130434e-05,
|
1391 |
+
"loss": 0.3021,
|
1392 |
+
"step": 2040
|
1393 |
+
},
|
1394 |
+
{
|
1395 |
+
"epoch": 7.17,
|
1396 |
+
"learning_rate": 8.847826086956521e-05,
|
1397 |
+
"loss": 0.3168,
|
1398 |
+
"step": 2050
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 7.2,
|
1402 |
+
"learning_rate": 8.739130434782608e-05,
|
1403 |
+
"loss": 0.2573,
|
1404 |
+
"step": 2060
|
1405 |
+
},
|
1406 |
+
{
|
1407 |
+
"epoch": 7.24,
|
1408 |
+
"learning_rate": 8.630434782608696e-05,
|
1409 |
+
"loss": 0.3107,
|
1410 |
+
"step": 2070
|
1411 |
+
},
|
1412 |
+
{
|
1413 |
+
"epoch": 7.27,
|
1414 |
+
"learning_rate": 8.521739130434782e-05,
|
1415 |
+
"loss": 0.2961,
|
1416 |
+
"step": 2080
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 7.31,
|
1420 |
+
"learning_rate": 8.413043478260869e-05,
|
1421 |
+
"loss": 0.3487,
|
1422 |
+
"step": 2090
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 7.34,
|
1426 |
+
"learning_rate": 8.304347826086954e-05,
|
1427 |
+
"loss": 0.337,
|
1428 |
+
"step": 2100
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 7.34,
|
1432 |
+
"eval_loss": 0.41391661763191223,
|
1433 |
+
"eval_runtime": 111.525,
|
1434 |
+
"eval_samples_per_second": 13.011,
|
1435 |
+
"eval_steps_per_second": 0.816,
|
1436 |
+
"step": 2100
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 7.38,
|
1440 |
+
"learning_rate": 8.195652173913043e-05,
|
1441 |
+
"loss": 0.3379,
|
1442 |
+
"step": 2110
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 7.41,
|
1446 |
+
"learning_rate": 8.08695652173913e-05,
|
1447 |
+
"loss": 0.3168,
|
1448 |
+
"step": 2120
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 7.45,
|
1452 |
+
"learning_rate": 7.978260869565217e-05,
|
1453 |
+
"loss": 0.3339,
|
1454 |
+
"step": 2130
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 7.48,
|
1458 |
+
"learning_rate": 7.869565217391304e-05,
|
1459 |
+
"loss": 0.3712,
|
1460 |
+
"step": 2140
|
1461 |
+
},
|
1462 |
+
{
|
1463 |
+
"epoch": 7.52,
|
1464 |
+
"learning_rate": 7.76086956521739e-05,
|
1465 |
+
"loss": 0.2807,
|
1466 |
+
"step": 2150
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 7.55,
|
1470 |
+
"learning_rate": 7.652173913043478e-05,
|
1471 |
+
"loss": 0.3018,
|
1472 |
+
"step": 2160
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 7.59,
|
1476 |
+
"learning_rate": 7.543478260869563e-05,
|
1477 |
+
"loss": 0.311,
|
1478 |
+
"step": 2170
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 7.62,
|
1482 |
+
"learning_rate": 7.434782608695652e-05,
|
1483 |
+
"loss": 0.309,
|
1484 |
+
"step": 2180
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 7.66,
|
1488 |
+
"learning_rate": 7.326086956521738e-05,
|
1489 |
+
"loss": 0.3175,
|
1490 |
+
"step": 2190
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 7.69,
|
1494 |
+
"learning_rate": 7.217391304347825e-05,
|
1495 |
+
"loss": 0.3045,
|
1496 |
+
"step": 2200
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 7.69,
|
1500 |
+
"eval_loss": 0.4140127897262573,
|
1501 |
+
"eval_runtime": 111.986,
|
1502 |
+
"eval_samples_per_second": 12.957,
|
1503 |
+
"eval_steps_per_second": 0.813,
|
1504 |
+
"step": 2200
|
1505 |
+
},
|
1506 |
+
{
|
1507 |
+
"epoch": 7.73,
|
1508 |
+
"learning_rate": 7.108695652173912e-05,
|
1509 |
+
"loss": 0.3065,
|
1510 |
+
"step": 2210
|
1511 |
+
},
|
1512 |
+
{
|
1513 |
+
"epoch": 7.76,
|
1514 |
+
"learning_rate": 7e-05,
|
1515 |
+
"loss": 0.2784,
|
1516 |
+
"step": 2220
|
1517 |
+
},
|
1518 |
+
{
|
1519 |
+
"epoch": 7.8,
|
1520 |
+
"learning_rate": 6.891304347826086e-05,
|
1521 |
+
"loss": 0.3367,
|
1522 |
+
"step": 2230
|
1523 |
+
},
|
1524 |
+
{
|
1525 |
+
"epoch": 7.83,
|
1526 |
+
"learning_rate": 6.782608695652173e-05,
|
1527 |
+
"loss": 0.3176,
|
1528 |
+
"step": 2240
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 7.87,
|
1532 |
+
"learning_rate": 6.67391304347826e-05,
|
1533 |
+
"loss": 0.3126,
|
1534 |
+
"step": 2250
|
1535 |
+
},
|
1536 |
+
{
|
1537 |
+
"epoch": 7.9,
|
1538 |
+
"learning_rate": 6.565217391304347e-05,
|
1539 |
+
"loss": 0.3586,
|
1540 |
+
"step": 2260
|
1541 |
+
},
|
1542 |
+
{
|
1543 |
+
"epoch": 7.94,
|
1544 |
+
"learning_rate": 6.456521739130434e-05,
|
1545 |
+
"loss": 0.3199,
|
1546 |
+
"step": 2270
|
1547 |
+
},
|
1548 |
+
{
|
1549 |
+
"epoch": 7.97,
|
1550 |
+
"learning_rate": 6.347826086956521e-05,
|
1551 |
+
"loss": 0.3302,
|
1552 |
+
"step": 2280
|
1553 |
+
},
|
1554 |
+
{
|
1555 |
+
"epoch": 8.01,
|
1556 |
+
"learning_rate": 6.239130434782608e-05,
|
1557 |
+
"loss": 0.346,
|
1558 |
+
"step": 2290
|
1559 |
+
},
|
1560 |
+
{
|
1561 |
+
"epoch": 8.04,
|
1562 |
+
"learning_rate": 6.130434782608695e-05,
|
1563 |
+
"loss": 0.2547,
|
1564 |
+
"step": 2300
|
1565 |
+
},
|
1566 |
+
{
|
1567 |
+
"epoch": 8.04,
|
1568 |
+
"eval_loss": 0.41661086678504944,
|
1569 |
+
"eval_runtime": 111.4943,
|
1570 |
+
"eval_samples_per_second": 13.014,
|
1571 |
+
"eval_steps_per_second": 0.816,
|
1572 |
+
"step": 2300
|
1573 |
+
},
|
1574 |
+
{
|
1575 |
+
"epoch": 8.08,
|
1576 |
+
"learning_rate": 6.021739130434782e-05,
|
1577 |
+
"loss": 0.2923,
|
1578 |
+
"step": 2310
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 8.11,
|
1582 |
+
"learning_rate": 5.913043478260869e-05,
|
1583 |
+
"loss": 0.2889,
|
1584 |
+
"step": 2320
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 8.15,
|
1588 |
+
"learning_rate": 5.804347826086956e-05,
|
1589 |
+
"loss": 0.2858,
|
1590 |
+
"step": 2330
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 8.18,
|
1594 |
+
"learning_rate": 5.695652173913043e-05,
|
1595 |
+
"loss": 0.3082,
|
1596 |
+
"step": 2340
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 8.22,
|
1600 |
+
"learning_rate": 5.58695652173913e-05,
|
1601 |
+
"loss": 0.2721,
|
1602 |
+
"step": 2350
|
1603 |
+
},
|
1604 |
+
{
|
1605 |
+
"epoch": 8.25,
|
1606 |
+
"learning_rate": 5.478260869565217e-05,
|
1607 |
+
"loss": 0.2911,
|
1608 |
+
"step": 2360
|
1609 |
+
},
|
1610 |
+
{
|
1611 |
+
"epoch": 8.29,
|
1612 |
+
"learning_rate": 5.3695652173913046e-05,
|
1613 |
+
"loss": 0.3015,
|
1614 |
+
"step": 2370
|
1615 |
+
},
|
1616 |
+
{
|
1617 |
+
"epoch": 8.32,
|
1618 |
+
"learning_rate": 5.260869565217391e-05,
|
1619 |
+
"loss": 0.3471,
|
1620 |
+
"step": 2380
|
1621 |
+
},
|
1622 |
+
{
|
1623 |
+
"epoch": 8.36,
|
1624 |
+
"learning_rate": 5.152173913043478e-05,
|
1625 |
+
"loss": 0.3053,
|
1626 |
+
"step": 2390
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 8.39,
|
1630 |
+
"learning_rate": 5.043478260869565e-05,
|
1631 |
+
"loss": 0.2584,
|
1632 |
+
"step": 2400
|
1633 |
+
},
|
1634 |
+
{
|
1635 |
+
"epoch": 8.39,
|
1636 |
+
"eval_loss": 0.408309668302536,
|
1637 |
+
"eval_runtime": 111.5385,
|
1638 |
+
"eval_samples_per_second": 13.009,
|
1639 |
+
"eval_steps_per_second": 0.816,
|
1640 |
+
"step": 2400
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 8.43,
|
1644 |
+
"learning_rate": 4.934782608695652e-05,
|
1645 |
+
"loss": 0.3029,
|
1646 |
+
"step": 2410
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 8.46,
|
1650 |
+
"learning_rate": 4.826086956521738e-05,
|
1651 |
+
"loss": 0.2933,
|
1652 |
+
"step": 2420
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 8.5,
|
1656 |
+
"learning_rate": 4.717391304347826e-05,
|
1657 |
+
"loss": 0.2807,
|
1658 |
+
"step": 2430
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 8.53,
|
1662 |
+
"learning_rate": 4.6086956521739126e-05,
|
1663 |
+
"loss": 0.2899,
|
1664 |
+
"step": 2440
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 8.57,
|
1668 |
+
"learning_rate": 4.4999999999999996e-05,
|
1669 |
+
"loss": 0.229,
|
1670 |
+
"step": 2450
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 8.6,
|
1674 |
+
"learning_rate": 4.3913043478260866e-05,
|
1675 |
+
"loss": 0.2919,
|
1676 |
+
"step": 2460
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 8.64,
|
1680 |
+
"learning_rate": 4.2826086956521735e-05,
|
1681 |
+
"loss": 0.2989,
|
1682 |
+
"step": 2470
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 8.67,
|
1686 |
+
"learning_rate": 4.1739130434782605e-05,
|
1687 |
+
"loss": 0.2919,
|
1688 |
+
"step": 2480
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 8.71,
|
1692 |
+
"learning_rate": 4.065217391304348e-05,
|
1693 |
+
"loss": 0.2988,
|
1694 |
+
"step": 2490
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 8.74,
|
1698 |
+
"learning_rate": 3.9565217391304344e-05,
|
1699 |
+
"loss": 0.2673,
|
1700 |
+
"step": 2500
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 8.74,
|
1704 |
+
"eval_loss": 0.39945822954177856,
|
1705 |
+
"eval_runtime": 111.4334,
|
1706 |
+
"eval_samples_per_second": 13.021,
|
1707 |
+
"eval_steps_per_second": 0.817,
|
1708 |
+
"step": 2500
|
1709 |
+
},
|
1710 |
+
{
|
1711 |
+
"epoch": 8.77,
|
1712 |
+
"learning_rate": 3.8478260869565214e-05,
|
1713 |
+
"loss": 0.2668,
|
1714 |
+
"step": 2510
|
1715 |
+
},
|
1716 |
+
{
|
1717 |
+
"epoch": 8.81,
|
1718 |
+
"learning_rate": 3.7391304347826084e-05,
|
1719 |
+
"loss": 0.3159,
|
1720 |
+
"step": 2520
|
1721 |
+
},
|
1722 |
+
{
|
1723 |
+
"epoch": 8.84,
|
1724 |
+
"learning_rate": 3.630434782608695e-05,
|
1725 |
+
"loss": 0.3037,
|
1726 |
+
"step": 2530
|
1727 |
+
},
|
1728 |
+
{
|
1729 |
+
"epoch": 8.88,
|
1730 |
+
"learning_rate": 3.521739130434782e-05,
|
1731 |
+
"loss": 0.2869,
|
1732 |
+
"step": 2540
|
1733 |
+
},
|
1734 |
+
{
|
1735 |
+
"epoch": 8.91,
|
1736 |
+
"learning_rate": 3.413043478260869e-05,
|
1737 |
+
"loss": 0.2792,
|
1738 |
+
"step": 2550
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 8.95,
|
1742 |
+
"learning_rate": 3.304347826086956e-05,
|
1743 |
+
"loss": 0.2939,
|
1744 |
+
"step": 2560
|
1745 |
+
},
|
1746 |
+
{
|
1747 |
+
"epoch": 8.98,
|
1748 |
+
"learning_rate": 3.195652173913043e-05,
|
1749 |
+
"loss": 0.2781,
|
1750 |
+
"step": 2570
|
1751 |
+
},
|
1752 |
+
{
|
1753 |
+
"epoch": 9.02,
|
1754 |
+
"learning_rate": 3.08695652173913e-05,
|
1755 |
+
"loss": 0.3154,
|
1756 |
+
"step": 2580
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 9.06,
|
1760 |
+
"learning_rate": 2.978260869565217e-05,
|
1761 |
+
"loss": 0.301,
|
1762 |
+
"step": 2590
|
1763 |
+
},
|
1764 |
+
{
|
1765 |
+
"epoch": 9.09,
|
1766 |
+
"learning_rate": 2.869565217391304e-05,
|
1767 |
+
"loss": 0.2765,
|
1768 |
+
"step": 2600
|
1769 |
+
},
|
1770 |
+
{
|
1771 |
+
"epoch": 9.09,
|
1772 |
+
"eval_loss": 0.4026023745536804,
|
1773 |
+
"eval_runtime": 111.3595,
|
1774 |
+
"eval_samples_per_second": 13.03,
|
1775 |
+
"eval_steps_per_second": 0.817,
|
1776 |
+
"step": 2600
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 9.13,
|
1780 |
+
"learning_rate": 2.760869565217391e-05,
|
1781 |
+
"loss": 0.2801,
|
1782 |
+
"step": 2610
|
1783 |
+
},
|
1784 |
+
{
|
1785 |
+
"epoch": 9.16,
|
1786 |
+
"learning_rate": 2.652173913043478e-05,
|
1787 |
+
"loss": 0.263,
|
1788 |
+
"step": 2620
|
1789 |
+
},
|
1790 |
+
{
|
1791 |
+
"epoch": 9.2,
|
1792 |
+
"learning_rate": 2.543478260869565e-05,
|
1793 |
+
"loss": 0.2586,
|
1794 |
+
"step": 2630
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 9.23,
|
1798 |
+
"learning_rate": 2.4347826086956516e-05,
|
1799 |
+
"loss": 0.2496,
|
1800 |
+
"step": 2640
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 9.27,
|
1804 |
+
"learning_rate": 2.3260869565217393e-05,
|
1805 |
+
"loss": 0.269,
|
1806 |
+
"step": 2650
|
1807 |
+
},
|
1808 |
+
{
|
1809 |
+
"epoch": 9.3,
|
1810 |
+
"learning_rate": 2.217391304347826e-05,
|
1811 |
+
"loss": 0.296,
|
1812 |
+
"step": 2660
|
1813 |
+
},
|
1814 |
+
{
|
1815 |
+
"epoch": 9.34,
|
1816 |
+
"learning_rate": 2.108695652173913e-05,
|
1817 |
+
"loss": 0.2531,
|
1818 |
+
"step": 2670
|
1819 |
+
},
|
1820 |
+
{
|
1821 |
+
"epoch": 9.37,
|
1822 |
+
"learning_rate": 1.9999999999999998e-05,
|
1823 |
+
"loss": 0.2871,
|
1824 |
+
"step": 2680
|
1825 |
+
},
|
1826 |
+
{
|
1827 |
+
"epoch": 9.4,
|
1828 |
+
"learning_rate": 1.8913043478260868e-05,
|
1829 |
+
"loss": 0.2616,
|
1830 |
+
"step": 2690
|
1831 |
+
},
|
1832 |
+
{
|
1833 |
+
"epoch": 9.44,
|
1834 |
+
"learning_rate": 1.7826086956521738e-05,
|
1835 |
+
"loss": 0.2453,
|
1836 |
+
"step": 2700
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 9.44,
|
1840 |
+
"eval_loss": 0.4079287052154541,
|
1841 |
+
"eval_runtime": 111.4001,
|
1842 |
+
"eval_samples_per_second": 13.025,
|
1843 |
+
"eval_steps_per_second": 0.817,
|
1844 |
+
"step": 2700
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 9.47,
|
1848 |
+
"learning_rate": 1.6739130434782607e-05,
|
1849 |
+
"loss": 0.3043,
|
1850 |
+
"step": 2710
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 9.51,
|
1854 |
+
"learning_rate": 1.5652173913043477e-05,
|
1855 |
+
"loss": 0.2297,
|
1856 |
+
"step": 2720
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 9.54,
|
1860 |
+
"learning_rate": 1.4565217391304347e-05,
|
1861 |
+
"loss": 0.2844,
|
1862 |
+
"step": 2730
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 9.58,
|
1866 |
+
"learning_rate": 1.3478260869565216e-05,
|
1867 |
+
"loss": 0.2485,
|
1868 |
+
"step": 2740
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 9.61,
|
1872 |
+
"learning_rate": 1.2391304347826086e-05,
|
1873 |
+
"loss": 0.2695,
|
1874 |
+
"step": 2750
|
1875 |
+
},
|
1876 |
+
{
|
1877 |
+
"epoch": 9.65,
|
1878 |
+
"learning_rate": 1.1304347826086956e-05,
|
1879 |
+
"loss": 0.3031,
|
1880 |
+
"step": 2760
|
1881 |
+
},
|
1882 |
+
{
|
1883 |
+
"epoch": 9.68,
|
1884 |
+
"learning_rate": 1.0217391304347825e-05,
|
1885 |
+
"loss": 0.2716,
|
1886 |
+
"step": 2770
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 9.72,
|
1890 |
+
"learning_rate": 9.130434782608695e-06,
|
1891 |
+
"loss": 0.2424,
|
1892 |
+
"step": 2780
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 9.75,
|
1896 |
+
"learning_rate": 8.043478260869565e-06,
|
1897 |
+
"loss": 0.2652,
|
1898 |
+
"step": 2790
|
1899 |
+
},
|
1900 |
+
{
|
1901 |
+
"epoch": 9.79,
|
1902 |
+
"learning_rate": 6.956521739130434e-06,
|
1903 |
+
"loss": 0.2883,
|
1904 |
+
"step": 2800
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 9.79,
|
1908 |
+
"eval_loss": 0.4038572311401367,
|
1909 |
+
"eval_runtime": 111.7946,
|
1910 |
+
"eval_samples_per_second": 12.979,
|
1911 |
+
"eval_steps_per_second": 0.814,
|
1912 |
+
"step": 2800
|
1913 |
+
},
|
1914 |
+
{
|
1915 |
+
"epoch": 9.82,
|
1916 |
+
"learning_rate": 5.869565217391305e-06,
|
1917 |
+
"loss": 0.2743,
|
1918 |
+
"step": 2810
|
1919 |
+
},
|
1920 |
+
{
|
1921 |
+
"epoch": 9.86,
|
1922 |
+
"learning_rate": 4.7826086956521735e-06,
|
1923 |
+
"loss": 0.2823,
|
1924 |
+
"step": 2820
|
1925 |
+
},
|
1926 |
+
{
|
1927 |
+
"epoch": 9.89,
|
1928 |
+
"learning_rate": 3.695652173913043e-06,
|
1929 |
+
"loss": 0.2766,
|
1930 |
+
"step": 2830
|
1931 |
+
},
|
1932 |
+
{
|
1933 |
+
"epoch": 9.93,
|
1934 |
+
"learning_rate": 2.608695652173913e-06,
|
1935 |
+
"loss": 0.2881,
|
1936 |
+
"step": 2840
|
1937 |
+
},
|
1938 |
+
{
|
1939 |
+
"epoch": 9.96,
|
1940 |
+
"learning_rate": 1.5217391304347823e-06,
|
1941 |
+
"loss": 0.3219,
|
1942 |
+
"step": 2850
|
1943 |
+
},
|
1944 |
+
{
|
1945 |
+
"epoch": 10.0,
|
1946 |
+
"learning_rate": 4.3478260869565214e-07,
|
1947 |
+
"loss": 0.294,
|
1948 |
+
"step": 2860
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 10.0,
|
1952 |
+
"step": 2860,
|
1953 |
+
"total_flos": 6.827433558323335e+18,
|
1954 |
+
"train_loss": 1.1227069218675574,
|
1955 |
+
"train_runtime": 9275.3367,
|
1956 |
+
"train_samples_per_second": 4.936,
|
1957 |
+
"train_steps_per_second": 0.308
|
1958 |
+
},
|
1959 |
+
{
|
1960 |
+
"epoch": 10.0,
|
1961 |
+
"eval_loss": 0.40599533915519714,
|
1962 |
+
"eval_runtime": 111.6061,
|
1963 |
+
"eval_samples_per_second": 13.001,
|
1964 |
+
"eval_steps_per_second": 0.815,
|
1965 |
+
"step": 2860
|
1966 |
+
}
|
1967 |
+
],
|
1968 |
+
"max_steps": 2860,
|
1969 |
+
"num_train_epochs": 10,
|
1970 |
+
"total_flos": 6.827433558323335e+18,
|
1971 |
+
"trial_name": null,
|
1972 |
+
"trial_params": null
|
1973 |
+
}
|
vocab.json
ADDED
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
" ": 50,
|
3 |
+
"'": 43,
|
4 |
+
"*": 28,
|
5 |
+
"[PAD]": 69,
|
6 |
+
"[UNK]": 68,
|
7 |
+
"a": 39,
|
8 |
+
"b": 44,
|
9 |
+
"c": 29,
|
10 |
+
"d": 4,
|
11 |
+
"e": 15,
|
12 |
+
"f": 40,
|
13 |
+
"g": 9,
|
14 |
+
"h": 41,
|
15 |
+
"i": 47,
|
16 |
+
"j": 7,
|
17 |
+
"k": 17,
|
18 |
+
"l": 0,
|
19 |
+
"m": 18,
|
20 |
+
"n": 48,
|
21 |
+
"o": 22,
|
22 |
+
"p": 34,
|
23 |
+
"q": 19,
|
24 |
+
"r": 36,
|
25 |
+
"s": 63,
|
26 |
+
"t": 46,
|
27 |
+
"u": 16,
|
28 |
+
"v": 59,
|
29 |
+
"w": 62,
|
30 |
+
"x": 55,
|
31 |
+
"y": 32,
|
32 |
+
"z": 65,
|
33 |
+
"æ": 56,
|
34 |
+
"ç": 52,
|
35 |
+
"ð": 37,
|
36 |
+
"ŋ": 5,
|
37 |
+
"ɑ": 42,
|
38 |
+
"ɔ": 54,
|
39 |
+
"ɕ": 14,
|
40 |
+
"ə": 64,
|
41 |
+
"ɛ": 38,
|
42 |
+
"ɝ": 33,
|
43 |
+
"ɪ": 1,
|
44 |
+
"ɯ": 20,
|
45 |
+
"ɲ": 49,
|
46 |
+
"ɴ": 26,
|
47 |
+
"ɸ": 31,
|
48 |
+
"ɾ": 23,
|
49 |
+
"ʃ": 58,
|
50 |
+
"ʊ": 57,
|
51 |
+
"ʌ": 45,
|
52 |
+
"ʑ": 27,
|
53 |
+
"ʒ": 30,
|
54 |
+
"ˈ": 51,
|
55 |
+
"ˌ": 3,
|
56 |
+
"ː": 8,
|
57 |
+
"͡": 61,
|
58 |
+
"θ": 60,
|
59 |
+
"―": 66,
|
60 |
+
"・": 11,
|
61 |
+
"凍": 25,
|
62 |
+
"牲": 6,
|
63 |
+
"犧": 10,
|
64 |
+
"繫": 35,
|
65 |
+
"雨": 12,
|
66 |
+
"-": 67,
|
67 |
+
"a": 53,
|
68 |
+
"h": 24,
|
69 |
+
"j": 13,
|
70 |
+
"「": 21,
|
71 |
+
"」": 2
|
72 |
+
}
|