Ethan Sim
commited on
Commit
•
12cd919
1
Parent(s):
848fa5f
stage best lsp fine adapt wce model
Browse files- config.json +62 -0
- generation_config.json +15 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- source.spm +0 -0
- special_tokens_map.json +5 -0
- target.spm +0 -0
- tokenizer_config.json +12 -0
- trainer_state.json +336 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-en-fr",
|
3 |
+
"_num_labels": 3,
|
4 |
+
"activation_dropout": 0.0,
|
5 |
+
"activation_function": "swish",
|
6 |
+
"add_bias_logits": false,
|
7 |
+
"add_final_layer_norm": false,
|
8 |
+
"architectures": [
|
9 |
+
"MarianMTModel"
|
10 |
+
],
|
11 |
+
"attention_dropout": 0.0,
|
12 |
+
"bad_words_ids": [
|
13 |
+
[
|
14 |
+
59513
|
15 |
+
]
|
16 |
+
],
|
17 |
+
"bos_token_id": 0,
|
18 |
+
"classif_dropout": 0.0,
|
19 |
+
"classifier_dropout": 0.0,
|
20 |
+
"d_model": 512,
|
21 |
+
"decoder_attention_heads": 8,
|
22 |
+
"decoder_ffn_dim": 2048,
|
23 |
+
"decoder_layerdrop": 0.0,
|
24 |
+
"decoder_layers": 6,
|
25 |
+
"decoder_start_token_id": 59513,
|
26 |
+
"decoder_vocab_size": 59514,
|
27 |
+
"dropout": 0.1,
|
28 |
+
"encoder_attention_heads": 8,
|
29 |
+
"encoder_ffn_dim": 2048,
|
30 |
+
"encoder_layerdrop": 0.0,
|
31 |
+
"encoder_layers": 6,
|
32 |
+
"eos_token_id": 0,
|
33 |
+
"forced_eos_token_id": 0,
|
34 |
+
"gradient_checkpointing": false,
|
35 |
+
"id2label": {
|
36 |
+
"0": "LABEL_0",
|
37 |
+
"1": "LABEL_1",
|
38 |
+
"2": "LABEL_2"
|
39 |
+
},
|
40 |
+
"init_std": 0.02,
|
41 |
+
"is_encoder_decoder": true,
|
42 |
+
"label2id": {
|
43 |
+
"LABEL_0": 0,
|
44 |
+
"LABEL_1": 1,
|
45 |
+
"LABEL_2": 2
|
46 |
+
},
|
47 |
+
"max_length": 512,
|
48 |
+
"max_position_embeddings": 512,
|
49 |
+
"model_type": "marian",
|
50 |
+
"normalize_before": false,
|
51 |
+
"normalize_embedding": false,
|
52 |
+
"num_beams": 4,
|
53 |
+
"num_hidden_layers": 6,
|
54 |
+
"pad_token_id": 59513,
|
55 |
+
"scale_embedding": true,
|
56 |
+
"share_encoder_decoder_embeddings": true,
|
57 |
+
"static_position_embeddings": true,
|
58 |
+
"torch_dtype": "float32",
|
59 |
+
"transformers_version": "4.28.0",
|
60 |
+
"use_cache": true,
|
61 |
+
"vocab_size": 59514
|
62 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bad_words_ids": [
|
3 |
+
[
|
4 |
+
59513
|
5 |
+
]
|
6 |
+
],
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"decoder_start_token_id": 59513,
|
9 |
+
"eos_token_id": 0,
|
10 |
+
"forced_eos_token_id": 0,
|
11 |
+
"max_length": 512,
|
12 |
+
"num_beams": 4,
|
13 |
+
"pad_token_id": 59513,
|
14 |
+
"transformers_version": "4.28.0"
|
15 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:95d3b1a11cb849fe5d459a046a2f15663b510d9abcf01061e6d896ee14520a4e
|
3 |
+
size 597026309
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4dc6fa5bc690b742d20bd466d6d82ac35b11325a9695e151bacbf329b3a3157f
|
3 |
+
size 298763205
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f3315841ff60adfa45417d1ec100c4d4993bcb371765c46294abbd0a98541b16
|
3 |
+
size 14575
|
scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7c6a06f9335e73c8ed93e221404b263810c470efc986283819aa2cfdfea7f275
|
3 |
+
size 557
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1378d1da724b1a77da62e79047ad80c84a64145612cd1941ad7ce9dd55ac66d8
|
3 |
+
size 627
|
source.spm
ADDED
Binary file (778 kB). View file
|
|
special_tokens_map.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eos_token": "</s>",
|
3 |
+
"pad_token": "<pad>",
|
4 |
+
"unk_token": "<unk>"
|
5 |
+
}
|
target.spm
ADDED
Binary file (802 kB). View file
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"clean_up_tokenization_spaces": true,
|
3 |
+
"eos_token": "</s>",
|
4 |
+
"model_max_length": 512,
|
5 |
+
"pad_token": "<pad>",
|
6 |
+
"separate_vocabs": false,
|
7 |
+
"source_lang": "en",
|
8 |
+
"sp_model_kwargs": {},
|
9 |
+
"target_lang": "fr",
|
10 |
+
"tokenizer_class": "MarianTokenizer",
|
11 |
+
"unk_token": "<unk>"
|
12 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,336 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 42.8215,
|
3 |
+
"best_model_checkpoint": "opus_base_lsp_adapt_wce_precision_1_ubweight_1.5/checkpoint-80000",
|
4 |
+
"epoch": 3.943217665615142,
|
5 |
+
"global_step": 80000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.2,
|
12 |
+
"learning_rate": 1.9753918572555207e-05,
|
13 |
+
"loss": 0.2154,
|
14 |
+
"step": 4000
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.2,
|
18 |
+
"eval_bleu": 34.2623,
|
19 |
+
"eval_gen_len": 41.1879,
|
20 |
+
"eval_loss": 0.10916412621736526,
|
21 |
+
"eval_runtime": 200.4936,
|
22 |
+
"eval_samples_per_second": 5.202,
|
23 |
+
"eval_steps_per_second": 0.165,
|
24 |
+
"step": 4000
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 0.39,
|
28 |
+
"learning_rate": 1.950746746845426e-05,
|
29 |
+
"loss": 0.1812,
|
30 |
+
"step": 8000
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.39,
|
34 |
+
"eval_bleu": 36.6395,
|
35 |
+
"eval_gen_len": 39.3643,
|
36 |
+
"eval_loss": 0.10679356753826141,
|
37 |
+
"eval_runtime": 186.1745,
|
38 |
+
"eval_samples_per_second": 5.602,
|
39 |
+
"eval_steps_per_second": 0.177,
|
40 |
+
"step": 8000
|
41 |
+
},
|
42 |
+
{
|
43 |
+
"epoch": 0.59,
|
44 |
+
"learning_rate": 1.9261016364353314e-05,
|
45 |
+
"loss": 0.1738,
|
46 |
+
"step": 12000
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 0.59,
|
50 |
+
"eval_bleu": 40.923,
|
51 |
+
"eval_gen_len": 35.1668,
|
52 |
+
"eval_loss": 0.10544949769973755,
|
53 |
+
"eval_runtime": 209.6195,
|
54 |
+
"eval_samples_per_second": 4.976,
|
55 |
+
"eval_steps_per_second": 0.157,
|
56 |
+
"step": 12000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.79,
|
60 |
+
"learning_rate": 1.9014565260252367e-05,
|
61 |
+
"loss": 0.168,
|
62 |
+
"step": 16000
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.79,
|
66 |
+
"eval_bleu": 40.9075,
|
67 |
+
"eval_gen_len": 35.9712,
|
68 |
+
"eval_loss": 0.10438334196805954,
|
69 |
+
"eval_runtime": 166.702,
|
70 |
+
"eval_samples_per_second": 6.257,
|
71 |
+
"eval_steps_per_second": 0.198,
|
72 |
+
"step": 16000
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.99,
|
76 |
+
"learning_rate": 1.876811415615142e-05,
|
77 |
+
"loss": 0.1655,
|
78 |
+
"step": 20000
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 0.99,
|
82 |
+
"eval_bleu": 41.7323,
|
83 |
+
"eval_gen_len": 34.9559,
|
84 |
+
"eval_loss": 0.10381767153739929,
|
85 |
+
"eval_runtime": 172.3391,
|
86 |
+
"eval_samples_per_second": 6.052,
|
87 |
+
"eval_steps_per_second": 0.191,
|
88 |
+
"step": 20000
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 1.18,
|
92 |
+
"learning_rate": 1.8521786277602524e-05,
|
93 |
+
"loss": 0.1534,
|
94 |
+
"step": 24000
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 1.18,
|
98 |
+
"eval_bleu": 41.8326,
|
99 |
+
"eval_gen_len": 35.2637,
|
100 |
+
"eval_loss": 0.10374099761247635,
|
101 |
+
"eval_runtime": 157.0866,
|
102 |
+
"eval_samples_per_second": 6.64,
|
103 |
+
"eval_steps_per_second": 0.21,
|
104 |
+
"step": 24000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.38,
|
108 |
+
"learning_rate": 1.8275520011829655e-05,
|
109 |
+
"loss": 0.1531,
|
110 |
+
"step": 28000
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.38,
|
114 |
+
"eval_bleu": 41.7809,
|
115 |
+
"eval_gen_len": 35.3979,
|
116 |
+
"eval_loss": 0.10289370268583298,
|
117 |
+
"eval_runtime": 156.7657,
|
118 |
+
"eval_samples_per_second": 6.653,
|
119 |
+
"eval_steps_per_second": 0.211,
|
120 |
+
"step": 28000
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 1.58,
|
124 |
+
"learning_rate": 1.8029130520504733e-05,
|
125 |
+
"loss": 0.1526,
|
126 |
+
"step": 32000
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 1.58,
|
130 |
+
"eval_bleu": 41.3462,
|
131 |
+
"eval_gen_len": 35.5944,
|
132 |
+
"eval_loss": 0.1023290827870369,
|
133 |
+
"eval_runtime": 150.6754,
|
134 |
+
"eval_samples_per_second": 6.922,
|
135 |
+
"eval_steps_per_second": 0.219,
|
136 |
+
"step": 32000
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 1.77,
|
140 |
+
"learning_rate": 1.7782741029179815e-05,
|
141 |
+
"loss": 0.1511,
|
142 |
+
"step": 36000
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 1.77,
|
146 |
+
"eval_bleu": 41.1111,
|
147 |
+
"eval_gen_len": 37.4535,
|
148 |
+
"eval_loss": 0.10196707397699356,
|
149 |
+
"eval_runtime": 127.7267,
|
150 |
+
"eval_samples_per_second": 8.166,
|
151 |
+
"eval_steps_per_second": 0.258,
|
152 |
+
"step": 36000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 1.97,
|
156 |
+
"learning_rate": 1.7536413150630914e-05,
|
157 |
+
"loss": 0.1503,
|
158 |
+
"step": 40000
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.97,
|
162 |
+
"eval_bleu": 42.3732,
|
163 |
+
"eval_gen_len": 34.6405,
|
164 |
+
"eval_loss": 0.1020611822605133,
|
165 |
+
"eval_runtime": 161.2219,
|
166 |
+
"eval_samples_per_second": 6.469,
|
167 |
+
"eval_steps_per_second": 0.205,
|
168 |
+
"step": 40000
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 2.17,
|
172 |
+
"learning_rate": 1.729008527208202e-05,
|
173 |
+
"loss": 0.1425,
|
174 |
+
"step": 44000
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 2.17,
|
178 |
+
"eval_bleu": 42.0327,
|
179 |
+
"eval_gen_len": 35.7593,
|
180 |
+
"eval_loss": 0.10245819389820099,
|
181 |
+
"eval_runtime": 141.648,
|
182 |
+
"eval_samples_per_second": 7.363,
|
183 |
+
"eval_steps_per_second": 0.233,
|
184 |
+
"step": 44000
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 2.37,
|
188 |
+
"learning_rate": 1.70436957807571e-05,
|
189 |
+
"loss": 0.1415,
|
190 |
+
"step": 48000
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 2.37,
|
194 |
+
"eval_bleu": 41.9016,
|
195 |
+
"eval_gen_len": 35.8274,
|
196 |
+
"eval_loss": 0.10226458311080933,
|
197 |
+
"eval_runtime": 136.9293,
|
198 |
+
"eval_samples_per_second": 7.617,
|
199 |
+
"eval_steps_per_second": 0.241,
|
200 |
+
"step": 48000
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.56,
|
204 |
+
"learning_rate": 1.6797367902208205e-05,
|
205 |
+
"loss": 0.1404,
|
206 |
+
"step": 52000
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.56,
|
210 |
+
"eval_bleu": 42.2252,
|
211 |
+
"eval_gen_len": 34.9367,
|
212 |
+
"eval_loss": 0.10195090621709824,
|
213 |
+
"eval_runtime": 152.282,
|
214 |
+
"eval_samples_per_second": 6.849,
|
215 |
+
"eval_steps_per_second": 0.217,
|
216 |
+
"step": 52000
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 2.76,
|
220 |
+
"learning_rate": 1.6551040023659308e-05,
|
221 |
+
"loss": 0.1406,
|
222 |
+
"step": 56000
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 2.76,
|
226 |
+
"eval_bleu": 42.1866,
|
227 |
+
"eval_gen_len": 35.0451,
|
228 |
+
"eval_loss": 0.10183615982532501,
|
229 |
+
"eval_runtime": 157.3058,
|
230 |
+
"eval_samples_per_second": 6.63,
|
231 |
+
"eval_steps_per_second": 0.21,
|
232 |
+
"step": 56000
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 2.96,
|
236 |
+
"learning_rate": 1.630471214511041e-05,
|
237 |
+
"loss": 0.1409,
|
238 |
+
"step": 60000
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 2.96,
|
242 |
+
"eval_bleu": 42.5067,
|
243 |
+
"eval_gen_len": 34.4919,
|
244 |
+
"eval_loss": 0.10114634037017822,
|
245 |
+
"eval_runtime": 120.4194,
|
246 |
+
"eval_samples_per_second": 8.661,
|
247 |
+
"eval_steps_per_second": 0.274,
|
248 |
+
"step": 60000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 3.15,
|
252 |
+
"learning_rate": 1.6058384266561514e-05,
|
253 |
+
"loss": 0.1344,
|
254 |
+
"step": 64000
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 3.15,
|
258 |
+
"eval_bleu": 42.1356,
|
259 |
+
"eval_gen_len": 33.9348,
|
260 |
+
"eval_loss": 0.10190291702747345,
|
261 |
+
"eval_runtime": 128.9076,
|
262 |
+
"eval_samples_per_second": 8.091,
|
263 |
+
"eval_steps_per_second": 0.256,
|
264 |
+
"step": 64000
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 3.35,
|
268 |
+
"learning_rate": 1.581205638801262e-05,
|
269 |
+
"loss": 0.1338,
|
270 |
+
"step": 68000
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 3.35,
|
274 |
+
"eval_bleu": 42.4556,
|
275 |
+
"eval_gen_len": 34.3605,
|
276 |
+
"eval_loss": 0.10205747187137604,
|
277 |
+
"eval_runtime": 124.6309,
|
278 |
+
"eval_samples_per_second": 8.369,
|
279 |
+
"eval_steps_per_second": 0.265,
|
280 |
+
"step": 68000
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 3.55,
|
284 |
+
"learning_rate": 1.55656668966877e-05,
|
285 |
+
"loss": 0.1326,
|
286 |
+
"step": 72000
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 3.55,
|
290 |
+
"eval_bleu": 42.1887,
|
291 |
+
"eval_gen_len": 34.8015,
|
292 |
+
"eval_loss": 0.1019921749830246,
|
293 |
+
"eval_runtime": 93.2673,
|
294 |
+
"eval_samples_per_second": 11.183,
|
295 |
+
"eval_steps_per_second": 0.354,
|
296 |
+
"step": 72000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 3.75,
|
300 |
+
"learning_rate": 1.5319339018138805e-05,
|
301 |
+
"loss": 0.133,
|
302 |
+
"step": 76000
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 3.75,
|
306 |
+
"eval_bleu": 42.457,
|
307 |
+
"eval_gen_len": 35.441,
|
308 |
+
"eval_loss": 0.10184559971094131,
|
309 |
+
"eval_runtime": 125.2762,
|
310 |
+
"eval_samples_per_second": 8.326,
|
311 |
+
"eval_steps_per_second": 0.263,
|
312 |
+
"step": 76000
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 3.94,
|
316 |
+
"learning_rate": 1.5073011139589906e-05,
|
317 |
+
"loss": 0.1338,
|
318 |
+
"step": 80000
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 3.94,
|
322 |
+
"eval_bleu": 42.8215,
|
323 |
+
"eval_gen_len": 35.1419,
|
324 |
+
"eval_loss": 0.10159751027822495,
|
325 |
+
"eval_runtime": 138.4069,
|
326 |
+
"eval_samples_per_second": 7.536,
|
327 |
+
"eval_steps_per_second": 0.238,
|
328 |
+
"step": 80000
|
329 |
+
}
|
330 |
+
],
|
331 |
+
"max_steps": 324608,
|
332 |
+
"num_train_epochs": 16,
|
333 |
+
"total_flos": 1.3796012702564352e+17,
|
334 |
+
"trial_name": null,
|
335 |
+
"trial_params": null
|
336 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ba45336c9f12b033cd935962b2af3b14ae1af95dfaa4241f509088de04c4ec00
|
3 |
+
size 3771
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|