Version 1
Browse files- config.json +76 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- trainer_state.json +510 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"apply_spec_augment": true,
|
5 |
+
"architectures": [
|
6 |
+
"Wav2Vec2ForCTC"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"bos_token_id": 1,
|
10 |
+
"conv_bias": true,
|
11 |
+
"conv_dim": [
|
12 |
+
512,
|
13 |
+
512,
|
14 |
+
512,
|
15 |
+
512,
|
16 |
+
512,
|
17 |
+
512,
|
18 |
+
512
|
19 |
+
],
|
20 |
+
"conv_kernel": [
|
21 |
+
10,
|
22 |
+
3,
|
23 |
+
3,
|
24 |
+
3,
|
25 |
+
3,
|
26 |
+
2,
|
27 |
+
2
|
28 |
+
],
|
29 |
+
"conv_stride": [
|
30 |
+
5,
|
31 |
+
2,
|
32 |
+
2,
|
33 |
+
2,
|
34 |
+
2,
|
35 |
+
2,
|
36 |
+
2
|
37 |
+
],
|
38 |
+
"ctc_loss_reduction": "mean",
|
39 |
+
"ctc_zero_infinity": false,
|
40 |
+
"do_stable_layer_norm": true,
|
41 |
+
"eos_token_id": 2,
|
42 |
+
"feat_extract_activation": "gelu",
|
43 |
+
"feat_extract_dropout": 0.0,
|
44 |
+
"feat_extract_norm": "layer",
|
45 |
+
"feat_proj_dropout": 0.0,
|
46 |
+
"final_dropout": 0.0,
|
47 |
+
"gradient_checkpointing": true,
|
48 |
+
"hidden_act": "gelu",
|
49 |
+
"hidden_dropout": 0.1,
|
50 |
+
"hidden_size": 1024,
|
51 |
+
"initializer_range": 0.02,
|
52 |
+
"intermediate_size": 4096,
|
53 |
+
"layer_norm_eps": 1e-05,
|
54 |
+
"layerdrop": 0.1,
|
55 |
+
"mask_channel_length": 10,
|
56 |
+
"mask_channel_min_space": 1,
|
57 |
+
"mask_channel_other": 0.0,
|
58 |
+
"mask_channel_prob": 0.0,
|
59 |
+
"mask_channel_selection": "static",
|
60 |
+
"mask_feature_length": 10,
|
61 |
+
"mask_feature_prob": 0.0,
|
62 |
+
"mask_time_length": 10,
|
63 |
+
"mask_time_min_space": 1,
|
64 |
+
"mask_time_other": 0.0,
|
65 |
+
"mask_time_prob": 0.05,
|
66 |
+
"mask_time_selection": "static",
|
67 |
+
"model_type": "wav2vec2",
|
68 |
+
"num_attention_heads": 16,
|
69 |
+
"num_conv_pos_embedding_groups": 16,
|
70 |
+
"num_conv_pos_embeddings": 128,
|
71 |
+
"num_feat_extract_layers": 7,
|
72 |
+
"num_hidden_layers": 24,
|
73 |
+
"pad_token_id": 41,
|
74 |
+
"transformers_version": "4.4.0.dev0",
|
75 |
+
"vocab_size": 42
|
76 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:39823d9bd704e711675ae48109d4cc2d266d58e12f8227be95133228c697e396
|
3 |
+
size 2490422279
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a28492891790caee8eda88933aded7a8bd21ed71bb97fcafdace6e675641c1be
|
3 |
+
size 1262106007
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6ecc9e980790d5f77be705a4f5cde16f657651bb91364b496dd824703d969308
|
3 |
+
size 623
|
trainer_state.json
ADDED
@@ -0,0 +1,510 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 8.104816696762272,
|
5 |
+
"global_step": 193000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.21,
|
12 |
+
"learning_rate": 0.0002955626928115623,
|
13 |
+
"loss": 1.1705,
|
14 |
+
"step": 5000
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.21,
|
18 |
+
"eval_loss": 0.22145646810531616,
|
19 |
+
"eval_runtime": 5082.7114,
|
20 |
+
"eval_samples_per_second": 4.02,
|
21 |
+
"step": 5000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.42,
|
25 |
+
"learning_rate": 0.0002892236825423657,
|
26 |
+
"loss": 0.2914,
|
27 |
+
"step": 10000
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.42,
|
31 |
+
"eval_loss": 0.19062571227550507,
|
32 |
+
"eval_runtime": 5086.3406,
|
33 |
+
"eval_samples_per_second": 4.018,
|
34 |
+
"step": 10000
|
35 |
+
},
|
36 |
+
{
|
37 |
+
"epoch": 0.63,
|
38 |
+
"learning_rate": 0.00028288467227316906,
|
39 |
+
"loss": 0.2617,
|
40 |
+
"step": 15000
|
41 |
+
},
|
42 |
+
{
|
43 |
+
"epoch": 0.63,
|
44 |
+
"eval_loss": 0.1784171611070633,
|
45 |
+
"eval_runtime": 4895.8348,
|
46 |
+
"eval_samples_per_second": 4.174,
|
47 |
+
"step": 15000
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 0.84,
|
51 |
+
"learning_rate": 0.0002765456620039724,
|
52 |
+
"loss": 0.2449,
|
53 |
+
"step": 20000
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 0.84,
|
57 |
+
"eval_loss": 0.16641439497470856,
|
58 |
+
"eval_runtime": 4945.3495,
|
59 |
+
"eval_samples_per_second": 4.132,
|
60 |
+
"step": 20000
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 1.05,
|
64 |
+
"learning_rate": 0.0002702066517347758,
|
65 |
+
"loss": 0.2264,
|
66 |
+
"step": 25000
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 1.05,
|
70 |
+
"eval_loss": 0.15650227665901184,
|
71 |
+
"eval_runtime": 4999.732,
|
72 |
+
"eval_samples_per_second": 4.087,
|
73 |
+
"step": 25000
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 1.26,
|
77 |
+
"learning_rate": 0.00026386764146557915,
|
78 |
+
"loss": 0.2093,
|
79 |
+
"step": 30000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 1.26,
|
83 |
+
"eval_loss": 0.15184776484966278,
|
84 |
+
"eval_runtime": 4877.0685,
|
85 |
+
"eval_samples_per_second": 4.19,
|
86 |
+
"step": 30000
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 1.47,
|
90 |
+
"learning_rate": 0.0002575286311963825,
|
91 |
+
"loss": 0.2003,
|
92 |
+
"step": 35000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 1.47,
|
96 |
+
"eval_loss": 0.14190027117729187,
|
97 |
+
"eval_runtime": 4879.5538,
|
98 |
+
"eval_samples_per_second": 4.188,
|
99 |
+
"step": 35000
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 1.68,
|
103 |
+
"learning_rate": 0.0002511896209271859,
|
104 |
+
"loss": 0.1962,
|
105 |
+
"step": 40000
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 1.68,
|
109 |
+
"eval_loss": 0.14596430957317352,
|
110 |
+
"eval_runtime": 4860.8654,
|
111 |
+
"eval_samples_per_second": 4.204,
|
112 |
+
"step": 40000
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 1.89,
|
116 |
+
"learning_rate": 0.00024485061065798925,
|
117 |
+
"loss": 0.19,
|
118 |
+
"step": 45000
|
119 |
+
},
|
120 |
+
{
|
121 |
+
"epoch": 1.89,
|
122 |
+
"eval_loss": 0.13479308784008026,
|
123 |
+
"eval_runtime": 4872.4496,
|
124 |
+
"eval_samples_per_second": 4.194,
|
125 |
+
"step": 45000
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 2.1,
|
129 |
+
"learning_rate": 0.00023851160038879262,
|
130 |
+
"loss": 0.1769,
|
131 |
+
"step": 50000
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 2.1,
|
135 |
+
"eval_loss": 0.13018357753753662,
|
136 |
+
"eval_runtime": 4872.3296,
|
137 |
+
"eval_samples_per_second": 4.194,
|
138 |
+
"step": 50000
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 2.31,
|
142 |
+
"learning_rate": 0.00023217259011959596,
|
143 |
+
"loss": 0.1674,
|
144 |
+
"step": 55000
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 2.31,
|
148 |
+
"eval_loss": 0.1303720772266388,
|
149 |
+
"eval_runtime": 4874.4649,
|
150 |
+
"eval_samples_per_second": 4.192,
|
151 |
+
"step": 55000
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 2.52,
|
155 |
+
"learning_rate": 0.00022583357985039935,
|
156 |
+
"loss": 0.1655,
|
157 |
+
"step": 60000
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 2.52,
|
161 |
+
"eval_loss": 0.12317115068435669,
|
162 |
+
"eval_runtime": 4882.2049,
|
163 |
+
"eval_samples_per_second": 4.186,
|
164 |
+
"step": 60000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 2.73,
|
168 |
+
"learning_rate": 0.00021949456958120271,
|
169 |
+
"loss": 0.1608,
|
170 |
+
"step": 65000
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 2.73,
|
174 |
+
"eval_loss": 0.12056649476289749,
|
175 |
+
"eval_runtime": 4877.3831,
|
176 |
+
"eval_samples_per_second": 4.19,
|
177 |
+
"step": 65000
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 2.94,
|
181 |
+
"learning_rate": 0.00021315555931200605,
|
182 |
+
"loss": 0.1565,
|
183 |
+
"step": 70000
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 2.94,
|
187 |
+
"eval_loss": 0.11486475169658661,
|
188 |
+
"eval_runtime": 4876.7317,
|
189 |
+
"eval_samples_per_second": 4.19,
|
190 |
+
"step": 70000
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 3.15,
|
194 |
+
"learning_rate": 0.00020681654904280945,
|
195 |
+
"loss": 0.146,
|
196 |
+
"step": 75000
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 3.15,
|
200 |
+
"eval_loss": 0.11829441785812378,
|
201 |
+
"eval_runtime": 4833.5153,
|
202 |
+
"eval_samples_per_second": 4.228,
|
203 |
+
"step": 75000
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 3.36,
|
207 |
+
"learning_rate": 0.00020047753877361279,
|
208 |
+
"loss": 0.1403,
|
209 |
+
"step": 80000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 3.36,
|
213 |
+
"eval_loss": 0.11143175512552261,
|
214 |
+
"eval_runtime": 4853.9814,
|
215 |
+
"eval_samples_per_second": 4.21,
|
216 |
+
"step": 80000
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 3.57,
|
220 |
+
"learning_rate": 0.00019413852850441618,
|
221 |
+
"loss": 0.1376,
|
222 |
+
"step": 85000
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 3.57,
|
226 |
+
"eval_loss": 0.11027190089225769,
|
227 |
+
"eval_runtime": 4858.6043,
|
228 |
+
"eval_samples_per_second": 4.206,
|
229 |
+
"step": 85000
|
230 |
+
},
|
231 |
+
{
|
232 |
+
"epoch": 3.78,
|
233 |
+
"learning_rate": 0.00018779951823521952,
|
234 |
+
"loss": 0.1337,
|
235 |
+
"step": 90000
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 3.78,
|
239 |
+
"eval_loss": 0.10872453451156616,
|
240 |
+
"eval_runtime": 4864.6388,
|
241 |
+
"eval_samples_per_second": 4.201,
|
242 |
+
"step": 90000
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 3.99,
|
246 |
+
"learning_rate": 0.00018146050796602288,
|
247 |
+
"loss": 0.1325,
|
248 |
+
"step": 95000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 3.99,
|
252 |
+
"eval_loss": 0.10718829929828644,
|
253 |
+
"eval_runtime": 4921.9134,
|
254 |
+
"eval_samples_per_second": 4.152,
|
255 |
+
"step": 95000
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 4.2,
|
259 |
+
"learning_rate": 0.00017512149769682625,
|
260 |
+
"loss": 0.1195,
|
261 |
+
"step": 100000
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 4.2,
|
265 |
+
"eval_loss": 0.10430513322353363,
|
266 |
+
"eval_runtime": 4864.6263,
|
267 |
+
"eval_samples_per_second": 4.201,
|
268 |
+
"step": 100000
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 4.41,
|
272 |
+
"learning_rate": 0.00016878248742762961,
|
273 |
+
"loss": 0.118,
|
274 |
+
"step": 105000
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 4.41,
|
278 |
+
"eval_loss": 0.1070966124534607,
|
279 |
+
"eval_runtime": 4879.2783,
|
280 |
+
"eval_samples_per_second": 4.188,
|
281 |
+
"step": 105000
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 4.62,
|
285 |
+
"learning_rate": 0.00016244347715843295,
|
286 |
+
"loss": 0.1173,
|
287 |
+
"step": 110000
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 4.62,
|
291 |
+
"eval_loss": 0.10433077067136765,
|
292 |
+
"eval_runtime": 4876.4984,
|
293 |
+
"eval_samples_per_second": 4.191,
|
294 |
+
"step": 110000
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 4.83,
|
298 |
+
"learning_rate": 0.00015610446688923635,
|
299 |
+
"loss": 0.115,
|
300 |
+
"step": 115000
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 4.83,
|
304 |
+
"eval_loss": 0.09682977199554443,
|
305 |
+
"eval_runtime": 4893.4652,
|
306 |
+
"eval_samples_per_second": 4.176,
|
307 |
+
"step": 115000
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 5.04,
|
311 |
+
"learning_rate": 0.0001497654566200397,
|
312 |
+
"loss": 0.1102,
|
313 |
+
"step": 120000
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 5.04,
|
317 |
+
"eval_loss": 0.09630288183689117,
|
318 |
+
"eval_runtime": 4914.2049,
|
319 |
+
"eval_samples_per_second": 4.158,
|
320 |
+
"step": 120000
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 5.25,
|
324 |
+
"learning_rate": 0.00014342644635084308,
|
325 |
+
"loss": 0.1019,
|
326 |
+
"step": 125000
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 5.25,
|
330 |
+
"eval_loss": 0.0918075293302536,
|
331 |
+
"eval_runtime": 4893.9499,
|
332 |
+
"eval_samples_per_second": 4.176,
|
333 |
+
"step": 125000
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 5.46,
|
337 |
+
"learning_rate": 0.00013708743608164644,
|
338 |
+
"loss": 0.1014,
|
339 |
+
"step": 130000
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 5.46,
|
343 |
+
"eval_loss": 0.09067174792289734,
|
344 |
+
"eval_runtime": 4891.4795,
|
345 |
+
"eval_samples_per_second": 4.178,
|
346 |
+
"step": 130000
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 5.67,
|
350 |
+
"learning_rate": 0.0001307484258124498,
|
351 |
+
"loss": 0.1,
|
352 |
+
"step": 135000
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 5.67,
|
356 |
+
"eval_loss": 0.08851899951696396,
|
357 |
+
"eval_runtime": 4884.1688,
|
358 |
+
"eval_samples_per_second": 4.184,
|
359 |
+
"step": 135000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 5.88,
|
363 |
+
"learning_rate": 0.00012440941554325318,
|
364 |
+
"loss": 0.0971,
|
365 |
+
"step": 140000
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 5.88,
|
369 |
+
"eval_loss": 0.08720648288726807,
|
370 |
+
"eval_runtime": 4896.4597,
|
371 |
+
"eval_samples_per_second": 4.173,
|
372 |
+
"step": 140000
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 6.09,
|
376 |
+
"learning_rate": 0.00011807040527405654,
|
377 |
+
"loss": 0.0921,
|
378 |
+
"step": 145000
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 6.09,
|
382 |
+
"eval_loss": 0.08666499704122543,
|
383 |
+
"eval_runtime": 5117.0953,
|
384 |
+
"eval_samples_per_second": 3.993,
|
385 |
+
"step": 145000
|
386 |
+
},
|
387 |
+
{
|
388 |
+
"epoch": 6.3,
|
389 |
+
"learning_rate": 0.00011173139500485991,
|
390 |
+
"loss": 0.0884,
|
391 |
+
"step": 150000
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 6.3,
|
395 |
+
"eval_loss": 0.0831904485821724,
|
396 |
+
"eval_runtime": 4879.0536,
|
397 |
+
"eval_samples_per_second": 4.188,
|
398 |
+
"step": 150000
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 6.51,
|
402 |
+
"learning_rate": 0.00010539238473566326,
|
403 |
+
"loss": 0.0864,
|
404 |
+
"step": 155000
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 6.51,
|
408 |
+
"eval_loss": 0.08337873965501785,
|
409 |
+
"eval_runtime": 4900.7475,
|
410 |
+
"eval_samples_per_second": 4.17,
|
411 |
+
"step": 155000
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 6.72,
|
415 |
+
"learning_rate": 9.905337446646663e-05,
|
416 |
+
"loss": 0.0861,
|
417 |
+
"step": 160000
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 6.72,
|
421 |
+
"eval_loss": 0.08155979961156845,
|
422 |
+
"eval_runtime": 4919.8246,
|
423 |
+
"eval_samples_per_second": 4.154,
|
424 |
+
"step": 160000
|
425 |
+
},
|
426 |
+
{
|
427 |
+
"epoch": 6.93,
|
428 |
+
"learning_rate": 9.271436419726999e-05,
|
429 |
+
"loss": 0.083,
|
430 |
+
"step": 165000
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 6.93,
|
434 |
+
"eval_loss": 0.08167865127325058,
|
435 |
+
"eval_runtime": 4921.4971,
|
436 |
+
"eval_samples_per_second": 4.152,
|
437 |
+
"step": 165000
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 7.14,
|
441 |
+
"learning_rate": 8.637535392807336e-05,
|
442 |
+
"loss": 0.0769,
|
443 |
+
"step": 170000
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 7.14,
|
447 |
+
"eval_loss": 0.0775604099035263,
|
448 |
+
"eval_runtime": 4893.4221,
|
449 |
+
"eval_samples_per_second": 4.176,
|
450 |
+
"step": 170000
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 7.35,
|
454 |
+
"learning_rate": 8.003634365887672e-05,
|
455 |
+
"loss": 0.0749,
|
456 |
+
"step": 175000
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 7.35,
|
460 |
+
"eval_loss": 0.07773936539888382,
|
461 |
+
"eval_runtime": 4914.0163,
|
462 |
+
"eval_samples_per_second": 4.159,
|
463 |
+
"step": 175000
|
464 |
+
},
|
465 |
+
{
|
466 |
+
"epoch": 7.56,
|
467 |
+
"learning_rate": 7.369733338968009e-05,
|
468 |
+
"loss": 0.0735,
|
469 |
+
"step": 180000
|
470 |
+
},
|
471 |
+
{
|
472 |
+
"epoch": 7.56,
|
473 |
+
"eval_loss": 0.07420430332422256,
|
474 |
+
"eval_runtime": 4934.0827,
|
475 |
+
"eval_samples_per_second": 4.142,
|
476 |
+
"step": 180000
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 7.77,
|
480 |
+
"learning_rate": 6.735832312048346e-05,
|
481 |
+
"loss": 0.0715,
|
482 |
+
"step": 185000
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 7.77,
|
486 |
+
"eval_loss": 0.07269106060266495,
|
487 |
+
"eval_runtime": 4925.1046,
|
488 |
+
"eval_samples_per_second": 4.149,
|
489 |
+
"step": 185000
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 7.98,
|
493 |
+
"learning_rate": 6.1019312851286814e-05,
|
494 |
+
"loss": 0.0702,
|
495 |
+
"step": 190000
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 7.98,
|
499 |
+
"eval_loss": 0.07183075696229935,
|
500 |
+
"eval_runtime": 4948.9933,
|
501 |
+
"eval_samples_per_second": 4.129,
|
502 |
+
"step": 190000
|
503 |
+
}
|
504 |
+
],
|
505 |
+
"max_steps": 238130,
|
506 |
+
"num_train_epochs": 10,
|
507 |
+
"total_flos": 5.387421756388246e+20,
|
508 |
+
"trial_name": null,
|
509 |
+
"trial_params": null
|
510 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ebe791c5dd04e87cd6dce636d0245fc7269b2260047be0dafb37e8e4f5eb0b6a
|
3 |
+
size 2287
|