Upload 13 files
Browse files- all_results.json +23 -0
- config.json +38 -0
- eval_results.json +10 -0
- generated_predictions.txt +0 -0
- predict_results.json +9 -0
- pytorch_model.bin +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +109 -0
- tokenizer_config.json +119 -0
- train_results.json +9 -0
- trainer_state.json +187 -0
- training_args.bin +3 -0
- vocab.json +0 -0
all_results.json
ADDED
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 40.0,
|
3 |
+
"eval_bleu": 11.5263,
|
4 |
+
"eval_gen_len": 40.7091,
|
5 |
+
"eval_loss": 3.6967673301696777,
|
6 |
+
"eval_runtime": 568.2025,
|
7 |
+
"eval_samples": 1564,
|
8 |
+
"eval_samples_per_second": 2.753,
|
9 |
+
"eval_steps_per_second": 0.276,
|
10 |
+
"predict_bleu": 10.0411,
|
11 |
+
"predict_gen_len": 42.7426,
|
12 |
+
"predict_loss": 3.71151065826416,
|
13 |
+
"predict_runtime": 601.9772,
|
14 |
+
"predict_samples": 1558,
|
15 |
+
"predict_samples_per_second": 2.588,
|
16 |
+
"predict_steps_per_second": 0.259,
|
17 |
+
"total_flos": 2.691983050788864e+16,
|
18 |
+
"train_loss": 0.3509185360711918,
|
19 |
+
"train_runtime": 12794.4297,
|
20 |
+
"train_samples": 3417,
|
21 |
+
"train_samples_per_second": 10.683,
|
22 |
+
"train_steps_per_second": 1.069
|
23 |
+
}
|
config.json
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/m2m100_418M",
|
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": 1024,
|
11 |
+
"decoder_attention_heads": 16,
|
12 |
+
"decoder_ffn_dim": 4096,
|
13 |
+
"decoder_layerdrop": 0.05,
|
14 |
+
"decoder_layers": 12,
|
15 |
+
"decoder_start_token_id": 2,
|
16 |
+
"dropout": 0.1,
|
17 |
+
"early_stopping": true,
|
18 |
+
"encoder_attention_heads": 16,
|
19 |
+
"encoder_ffn_dim": 4096,
|
20 |
+
"encoder_layerdrop": 0.05,
|
21 |
+
"encoder_layers": 12,
|
22 |
+
"eos_token_id": 2,
|
23 |
+
"forced_bos_token_id": 128028,
|
24 |
+
"gradient_checkpointing": false,
|
25 |
+
"init_std": 0.02,
|
26 |
+
"is_encoder_decoder": true,
|
27 |
+
"max_length": 200,
|
28 |
+
"max_position_embeddings": 1024,
|
29 |
+
"model_type": "m2m_100",
|
30 |
+
"num_beams": 5,
|
31 |
+
"num_hidden_layers": 12,
|
32 |
+
"pad_token_id": 1,
|
33 |
+
"scale_embedding": true,
|
34 |
+
"torch_dtype": "float32",
|
35 |
+
"transformers_version": "4.24.0",
|
36 |
+
"use_cache": true,
|
37 |
+
"vocab_size": 128112
|
38 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 40.0,
|
3 |
+
"eval_bleu": 11.5263,
|
4 |
+
"eval_gen_len": 40.7091,
|
5 |
+
"eval_loss": 3.6967673301696777,
|
6 |
+
"eval_runtime": 568.2025,
|
7 |
+
"eval_samples": 1564,
|
8 |
+
"eval_samples_per_second": 2.753,
|
9 |
+
"eval_steps_per_second": 0.276
|
10 |
+
}
|
generated_predictions.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
predict_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"predict_bleu": 10.0411,
|
3 |
+
"predict_gen_len": 42.7426,
|
4 |
+
"predict_loss": 3.71151065826416,
|
5 |
+
"predict_runtime": 601.9772,
|
6 |
+
"predict_samples": 1558,
|
7 |
+
"predict_samples_per_second": 2.588,
|
8 |
+
"predict_steps_per_second": 0.259
|
9 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8d7856e375a9ba202931637e0cb367e7a9dae2ee5c51b459676b9fcfc58d6c76
|
3 |
+
size 1935792071
|
sentencepiece.bpe.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d8f7c76ed2a5e0822be39f0a4f95a55eb19c78f4593ce609e2edbc2aea4d380a
|
3 |
+
size 2423393
|
special_tokens_map.json
ADDED
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"__af__",
|
4 |
+
"__am__",
|
5 |
+
"__ar__",
|
6 |
+
"__ast__",
|
7 |
+
"__az__",
|
8 |
+
"__ba__",
|
9 |
+
"__be__",
|
10 |
+
"__bg__",
|
11 |
+
"__bn__",
|
12 |
+
"__br__",
|
13 |
+
"__bs__",
|
14 |
+
"__ca__",
|
15 |
+
"__ceb__",
|
16 |
+
"__cs__",
|
17 |
+
"__cy__",
|
18 |
+
"__da__",
|
19 |
+
"__de__",
|
20 |
+
"__el__",
|
21 |
+
"__en__",
|
22 |
+
"__es__",
|
23 |
+
"__et__",
|
24 |
+
"__fa__",
|
25 |
+
"__ff__",
|
26 |
+
"__fi__",
|
27 |
+
"__fr__",
|
28 |
+
"__fy__",
|
29 |
+
"__ga__",
|
30 |
+
"__gd__",
|
31 |
+
"__gl__",
|
32 |
+
"__gu__",
|
33 |
+
"__ha__",
|
34 |
+
"__he__",
|
35 |
+
"__hi__",
|
36 |
+
"__hr__",
|
37 |
+
"__ht__",
|
38 |
+
"__hu__",
|
39 |
+
"__hy__",
|
40 |
+
"__id__",
|
41 |
+
"__ig__",
|
42 |
+
"__ilo__",
|
43 |
+
"__is__",
|
44 |
+
"__it__",
|
45 |
+
"__ja__",
|
46 |
+
"__jv__",
|
47 |
+
"__ka__",
|
48 |
+
"__kk__",
|
49 |
+
"__km__",
|
50 |
+
"__kn__",
|
51 |
+
"__ko__",
|
52 |
+
"__lb__",
|
53 |
+
"__lg__",
|
54 |
+
"__ln__",
|
55 |
+
"__lo__",
|
56 |
+
"__lt__",
|
57 |
+
"__lv__",
|
58 |
+
"__mg__",
|
59 |
+
"__mk__",
|
60 |
+
"__ml__",
|
61 |
+
"__mn__",
|
62 |
+
"__mr__",
|
63 |
+
"__ms__",
|
64 |
+
"__my__",
|
65 |
+
"__ne__",
|
66 |
+
"__nl__",
|
67 |
+
"__no__",
|
68 |
+
"__ns__",
|
69 |
+
"__oc__",
|
70 |
+
"__or__",
|
71 |
+
"__pa__",
|
72 |
+
"__pl__",
|
73 |
+
"__ps__",
|
74 |
+
"__pt__",
|
75 |
+
"__ro__",
|
76 |
+
"__ru__",
|
77 |
+
"__sd__",
|
78 |
+
"__si__",
|
79 |
+
"__sk__",
|
80 |
+
"__sl__",
|
81 |
+
"__so__",
|
82 |
+
"__sq__",
|
83 |
+
"__sr__",
|
84 |
+
"__ss__",
|
85 |
+
"__su__",
|
86 |
+
"__sv__",
|
87 |
+
"__sw__",
|
88 |
+
"__ta__",
|
89 |
+
"__th__",
|
90 |
+
"__tl__",
|
91 |
+
"__tn__",
|
92 |
+
"__tr__",
|
93 |
+
"__uk__",
|
94 |
+
"__ur__",
|
95 |
+
"__uz__",
|
96 |
+
"__vi__",
|
97 |
+
"__wo__",
|
98 |
+
"__xh__",
|
99 |
+
"__yi__",
|
100 |
+
"__yo__",
|
101 |
+
"__zh__",
|
102 |
+
"__zu__"
|
103 |
+
],
|
104 |
+
"bos_token": "<s>",
|
105 |
+
"eos_token": "</s>",
|
106 |
+
"pad_token": "<pad>",
|
107 |
+
"sep_token": "</s>",
|
108 |
+
"unk_token": "<unk>"
|
109 |
+
}
|
tokenizer_config.json
ADDED
@@ -0,0 +1,119 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"__af__",
|
4 |
+
"__am__",
|
5 |
+
"__ar__",
|
6 |
+
"__ast__",
|
7 |
+
"__az__",
|
8 |
+
"__ba__",
|
9 |
+
"__be__",
|
10 |
+
"__bg__",
|
11 |
+
"__bn__",
|
12 |
+
"__br__",
|
13 |
+
"__bs__",
|
14 |
+
"__ca__",
|
15 |
+
"__ceb__",
|
16 |
+
"__cs__",
|
17 |
+
"__cy__",
|
18 |
+
"__da__",
|
19 |
+
"__de__",
|
20 |
+
"__el__",
|
21 |
+
"__en__",
|
22 |
+
"__es__",
|
23 |
+
"__et__",
|
24 |
+
"__fa__",
|
25 |
+
"__ff__",
|
26 |
+
"__fi__",
|
27 |
+
"__fr__",
|
28 |
+
"__fy__",
|
29 |
+
"__ga__",
|
30 |
+
"__gd__",
|
31 |
+
"__gl__",
|
32 |
+
"__gu__",
|
33 |
+
"__ha__",
|
34 |
+
"__he__",
|
35 |
+
"__hi__",
|
36 |
+
"__hr__",
|
37 |
+
"__ht__",
|
38 |
+
"__hu__",
|
39 |
+
"__hy__",
|
40 |
+
"__id__",
|
41 |
+
"__ig__",
|
42 |
+
"__ilo__",
|
43 |
+
"__is__",
|
44 |
+
"__it__",
|
45 |
+
"__ja__",
|
46 |
+
"__jv__",
|
47 |
+
"__ka__",
|
48 |
+
"__kk__",
|
49 |
+
"__km__",
|
50 |
+
"__kn__",
|
51 |
+
"__ko__",
|
52 |
+
"__lb__",
|
53 |
+
"__lg__",
|
54 |
+
"__ln__",
|
55 |
+
"__lo__",
|
56 |
+
"__lt__",
|
57 |
+
"__lv__",
|
58 |
+
"__mg__",
|
59 |
+
"__mk__",
|
60 |
+
"__ml__",
|
61 |
+
"__mn__",
|
62 |
+
"__mr__",
|
63 |
+
"__ms__",
|
64 |
+
"__my__",
|
65 |
+
"__ne__",
|
66 |
+
"__nl__",
|
67 |
+
"__no__",
|
68 |
+
"__ns__",
|
69 |
+
"__oc__",
|
70 |
+
"__or__",
|
71 |
+
"__pa__",
|
72 |
+
"__pl__",
|
73 |
+
"__ps__",
|
74 |
+
"__pt__",
|
75 |
+
"__ro__",
|
76 |
+
"__ru__",
|
77 |
+
"__sd__",
|
78 |
+
"__si__",
|
79 |
+
"__sk__",
|
80 |
+
"__sl__",
|
81 |
+
"__so__",
|
82 |
+
"__sq__",
|
83 |
+
"__sr__",
|
84 |
+
"__ss__",
|
85 |
+
"__su__",
|
86 |
+
"__sv__",
|
87 |
+
"__sw__",
|
88 |
+
"__ta__",
|
89 |
+
"__th__",
|
90 |
+
"__tl__",
|
91 |
+
"__tn__",
|
92 |
+
"__tr__",
|
93 |
+
"__uk__",
|
94 |
+
"__ur__",
|
95 |
+
"__uz__",
|
96 |
+
"__vi__",
|
97 |
+
"__wo__",
|
98 |
+
"__xh__",
|
99 |
+
"__yi__",
|
100 |
+
"__yo__",
|
101 |
+
"__zh__",
|
102 |
+
"__zu__"
|
103 |
+
],
|
104 |
+
"bos_token": "<s>",
|
105 |
+
"eos_token": "</s>",
|
106 |
+
"language_codes": "m2m100",
|
107 |
+
"model_max_length": 1024,
|
108 |
+
"name_or_path": "facebook/m2m100_418M",
|
109 |
+
"num_madeup_words": 8,
|
110 |
+
"pad_token": "<pad>",
|
111 |
+
"sep_token": "</s>",
|
112 |
+
"sp_model_kwargs": {},
|
113 |
+
"special_tokens_map_file": "m2m_100_1.2B_v2/special_tokens_map.json",
|
114 |
+
"src_lang": null,
|
115 |
+
"tgt_lang": null,
|
116 |
+
"tokenizer_class": "M2M100Tokenizer",
|
117 |
+
"tokenizer_file": null,
|
118 |
+
"unk_token": "<unk>"
|
119 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 40.0,
|
3 |
+
"total_flos": 2.691983050788864e+16,
|
4 |
+
"train_loss": 0.3509185360711918,
|
5 |
+
"train_runtime": 12794.4297,
|
6 |
+
"train_samples": 3417,
|
7 |
+
"train_samples_per_second": 10.683,
|
8 |
+
"train_steps_per_second": 1.069
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,187 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 40.0,
|
5 |
+
"global_step": 13680,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 1.46,
|
12 |
+
"learning_rate": 4.817251461988304e-05,
|
13 |
+
"loss": 2.8952,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 2.92,
|
18 |
+
"learning_rate": 4.634502923976608e-05,
|
19 |
+
"loss": 2.0346,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 4.39,
|
24 |
+
"learning_rate": 4.451754385964912e-05,
|
25 |
+
"loss": 1.4152,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 5.85,
|
30 |
+
"learning_rate": 4.269005847953216e-05,
|
31 |
+
"loss": 1.009,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 7.31,
|
36 |
+
"learning_rate": 4.086257309941521e-05,
|
37 |
+
"loss": 0.6471,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 8.77,
|
42 |
+
"learning_rate": 3.9035087719298244e-05,
|
43 |
+
"loss": 0.4255,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 10.23,
|
48 |
+
"learning_rate": 3.7207602339181284e-05,
|
49 |
+
"loss": 0.2694,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 11.7,
|
54 |
+
"learning_rate": 3.538011695906433e-05,
|
55 |
+
"loss": 0.1837,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 13.16,
|
60 |
+
"learning_rate": 3.355263157894737e-05,
|
61 |
+
"loss": 0.1477,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 14.62,
|
66 |
+
"learning_rate": 3.172514619883041e-05,
|
67 |
+
"loss": 0.0993,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 16.08,
|
72 |
+
"learning_rate": 2.989766081871345e-05,
|
73 |
+
"loss": 0.0836,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 17.54,
|
78 |
+
"learning_rate": 2.8070175438596492e-05,
|
79 |
+
"loss": 0.0599,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 19.01,
|
84 |
+
"learning_rate": 2.6242690058479535e-05,
|
85 |
+
"loss": 0.0526,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 20.47,
|
90 |
+
"learning_rate": 2.4415204678362576e-05,
|
91 |
+
"loss": 0.0451,
|
92 |
+
"step": 7000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 21.93,
|
96 |
+
"learning_rate": 2.2587719298245616e-05,
|
97 |
+
"loss": 0.038,
|
98 |
+
"step": 7500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 23.39,
|
102 |
+
"learning_rate": 2.0760233918128656e-05,
|
103 |
+
"loss": 0.0299,
|
104 |
+
"step": 8000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 24.85,
|
108 |
+
"learning_rate": 1.8932748538011696e-05,
|
109 |
+
"loss": 0.036,
|
110 |
+
"step": 8500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 26.32,
|
114 |
+
"learning_rate": 1.7105263157894737e-05,
|
115 |
+
"loss": 0.0213,
|
116 |
+
"step": 9000
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 27.78,
|
120 |
+
"learning_rate": 1.527777777777778e-05,
|
121 |
+
"loss": 0.0195,
|
122 |
+
"step": 9500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 29.24,
|
126 |
+
"learning_rate": 1.3450292397660819e-05,
|
127 |
+
"loss": 0.0168,
|
128 |
+
"step": 10000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 30.7,
|
132 |
+
"learning_rate": 1.162280701754386e-05,
|
133 |
+
"loss": 0.0137,
|
134 |
+
"step": 10500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 32.16,
|
138 |
+
"learning_rate": 9.795321637426901e-06,
|
139 |
+
"loss": 0.0132,
|
140 |
+
"step": 11000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 33.63,
|
144 |
+
"learning_rate": 7.967836257309943e-06,
|
145 |
+
"loss": 0.0103,
|
146 |
+
"step": 11500
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 35.09,
|
150 |
+
"learning_rate": 6.140350877192982e-06,
|
151 |
+
"loss": 0.0085,
|
152 |
+
"step": 12000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 36.55,
|
156 |
+
"learning_rate": 4.312865497076023e-06,
|
157 |
+
"loss": 0.0085,
|
158 |
+
"step": 12500
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 38.01,
|
162 |
+
"learning_rate": 2.4853801169590643e-06,
|
163 |
+
"loss": 0.0083,
|
164 |
+
"step": 13000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 39.47,
|
168 |
+
"learning_rate": 6.578947368421053e-07,
|
169 |
+
"loss": 0.0072,
|
170 |
+
"step": 13500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 40.0,
|
174 |
+
"step": 13680,
|
175 |
+
"total_flos": 2.691983050788864e+16,
|
176 |
+
"train_loss": 0.3509185360711918,
|
177 |
+
"train_runtime": 12794.4297,
|
178 |
+
"train_samples_per_second": 10.683,
|
179 |
+
"train_steps_per_second": 1.069
|
180 |
+
}
|
181 |
+
],
|
182 |
+
"max_steps": 13680,
|
183 |
+
"num_train_epochs": 40,
|
184 |
+
"total_flos": 2.691983050788864e+16,
|
185 |
+
"trial_name": null,
|
186 |
+
"trial_params": null
|
187 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4b11c3b68bb8bfafe01986852c8d83013412c74dc619083615d5eb3bcfb96696
|
3 |
+
size 3503
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|