logoyazilim
commited on
Commit
•
7a6be3c
1
Parent(s):
bc74061
commit files to HF hub
Browse files- config.json +31 -0
- hyper_params.json +8 -0
- pytorch_model.bin +3 -0
- trainer_state.json +685 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "logoyazilim/qna_model_0000_1",
|
3 |
+
"architectures": [
|
4 |
+
"MT5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"d_ff": 1024,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 512,
|
9 |
+
"decoder_start_token_id": 0,
|
10 |
+
"dense_act_fn": "gelu_new",
|
11 |
+
"dropout_rate": 0.1,
|
12 |
+
"eos_token_id": 1,
|
13 |
+
"feed_forward_proj": "gated-gelu",
|
14 |
+
"initializer_factor": 1.0,
|
15 |
+
"is_encoder_decoder": true,
|
16 |
+
"is_gated_act": true,
|
17 |
+
"layer_norm_epsilon": 1e-06,
|
18 |
+
"model_type": "mt5",
|
19 |
+
"num_decoder_layers": 8,
|
20 |
+
"num_heads": 6,
|
21 |
+
"num_layers": 8,
|
22 |
+
"pad_token_id": 0,
|
23 |
+
"relative_attention_max_distance": 128,
|
24 |
+
"relative_attention_num_buckets": 32,
|
25 |
+
"tie_word_embeddings": false,
|
26 |
+
"tokenizer_class": "T5Tokenizer",
|
27 |
+
"torch_dtype": "float32",
|
28 |
+
"transformers_version": "4.22.2",
|
29 |
+
"use_cache": true,
|
30 |
+
"vocab_size": 250102
|
31 |
+
}
|
hyper_params.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"per_device_train_batch_size": 8,
|
3 |
+
"per_device_eval_batch_size": 8,
|
4 |
+
"gradient_accumulation_steps": 4,
|
5 |
+
"learning_rate": 0.0005,
|
6 |
+
"num_train_epochs": 8,
|
7 |
+
"max_steps": -1
|
8 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:15b41170dceb5ff10d9f3d7e556578d49d4516cfeb18046b185be4459d67e055
|
3 |
+
size 1200729797
|
trainer_state.json
ADDED
@@ -0,0 +1,685 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 7.999927855133107,
|
5 |
+
"global_step": 55440,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.07,
|
12 |
+
"learning_rate": 0.0004954906204906205,
|
13 |
+
"loss": 0.5606,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.14,
|
18 |
+
"learning_rate": 0.000490981240981241,
|
19 |
+
"loss": 0.4359,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.22,
|
24 |
+
"learning_rate": 0.00048647186147186144,
|
25 |
+
"loss": 0.3852,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.29,
|
30 |
+
"learning_rate": 0.000481962481962482,
|
31 |
+
"loss": 0.3778,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.36,
|
36 |
+
"learning_rate": 0.00047745310245310245,
|
37 |
+
"loss": 0.3523,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.43,
|
42 |
+
"learning_rate": 0.0004729437229437229,
|
43 |
+
"loss": 0.3319,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.51,
|
48 |
+
"learning_rate": 0.00046843434343434346,
|
49 |
+
"loss": 0.3153,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.58,
|
54 |
+
"learning_rate": 0.00046392496392496394,
|
55 |
+
"loss": 0.3027,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.65,
|
60 |
+
"learning_rate": 0.0004594155844155844,
|
61 |
+
"loss": 0.3063,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.72,
|
66 |
+
"learning_rate": 0.00045490620490620494,
|
67 |
+
"loss": 0.292,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.79,
|
72 |
+
"learning_rate": 0.0004503968253968254,
|
73 |
+
"loss": 0.2802,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.87,
|
78 |
+
"learning_rate": 0.0004458874458874459,
|
79 |
+
"loss": 0.2705,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.94,
|
84 |
+
"learning_rate": 0.0004413780663780664,
|
85 |
+
"loss": 0.2669,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 1.01,
|
90 |
+
"learning_rate": 0.00043686868686868685,
|
91 |
+
"loss": 0.2616,
|
92 |
+
"step": 7000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 1.08,
|
96 |
+
"learning_rate": 0.00043235930735930733,
|
97 |
+
"loss": 0.2486,
|
98 |
+
"step": 7500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 1.15,
|
102 |
+
"learning_rate": 0.00042784992784992786,
|
103 |
+
"loss": 0.2432,
|
104 |
+
"step": 8000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.23,
|
108 |
+
"learning_rate": 0.00042334054834054834,
|
109 |
+
"loss": 0.2413,
|
110 |
+
"step": 8500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.3,
|
114 |
+
"learning_rate": 0.0004188311688311688,
|
115 |
+
"loss": 0.2479,
|
116 |
+
"step": 9000
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 1.37,
|
120 |
+
"learning_rate": 0.00041432178932178935,
|
121 |
+
"loss": 0.2344,
|
122 |
+
"step": 9500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 1.44,
|
126 |
+
"learning_rate": 0.0004098124098124098,
|
127 |
+
"loss": 0.2363,
|
128 |
+
"step": 10000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.52,
|
132 |
+
"learning_rate": 0.0004053030303030303,
|
133 |
+
"loss": 0.2328,
|
134 |
+
"step": 10500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.59,
|
138 |
+
"learning_rate": 0.00040079365079365083,
|
139 |
+
"loss": 0.237,
|
140 |
+
"step": 11000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.66,
|
144 |
+
"learning_rate": 0.0003962842712842713,
|
145 |
+
"loss": 0.2284,
|
146 |
+
"step": 11500
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 1.73,
|
150 |
+
"learning_rate": 0.0003917748917748918,
|
151 |
+
"loss": 0.2286,
|
152 |
+
"step": 12000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 1.8,
|
156 |
+
"learning_rate": 0.00038726551226551226,
|
157 |
+
"loss": 0.222,
|
158 |
+
"step": 12500
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.88,
|
162 |
+
"learning_rate": 0.00038275613275613274,
|
163 |
+
"loss": 0.2144,
|
164 |
+
"step": 13000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 1.95,
|
168 |
+
"learning_rate": 0.0003782467532467532,
|
169 |
+
"loss": 0.2272,
|
170 |
+
"step": 13500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 2.02,
|
174 |
+
"learning_rate": 0.00037373737373737375,
|
175 |
+
"loss": 0.2107,
|
176 |
+
"step": 14000
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 2.09,
|
180 |
+
"learning_rate": 0.0003692279942279942,
|
181 |
+
"loss": 0.2054,
|
182 |
+
"step": 14500
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 2.16,
|
186 |
+
"learning_rate": 0.0003647186147186147,
|
187 |
+
"loss": 0.1953,
|
188 |
+
"step": 15000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 2.24,
|
192 |
+
"learning_rate": 0.00036020923520923523,
|
193 |
+
"loss": 0.1955,
|
194 |
+
"step": 15500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 2.31,
|
198 |
+
"learning_rate": 0.0003556998556998557,
|
199 |
+
"loss": 0.2012,
|
200 |
+
"step": 16000
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.38,
|
204 |
+
"learning_rate": 0.0003511904761904762,
|
205 |
+
"loss": 0.1938,
|
206 |
+
"step": 16500
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.45,
|
210 |
+
"learning_rate": 0.0003466810966810967,
|
211 |
+
"loss": 0.1932,
|
212 |
+
"step": 17000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 2.53,
|
216 |
+
"learning_rate": 0.0003421717171717172,
|
217 |
+
"loss": 0.1944,
|
218 |
+
"step": 17500
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.6,
|
222 |
+
"learning_rate": 0.00033766233766233767,
|
223 |
+
"loss": 0.1967,
|
224 |
+
"step": 18000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.67,
|
228 |
+
"learning_rate": 0.00033315295815295815,
|
229 |
+
"loss": 0.1955,
|
230 |
+
"step": 18500
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 2.74,
|
234 |
+
"learning_rate": 0.0003286435786435786,
|
235 |
+
"loss": 0.1906,
|
236 |
+
"step": 19000
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 2.81,
|
240 |
+
"learning_rate": 0.0003241341991341991,
|
241 |
+
"loss": 0.1894,
|
242 |
+
"step": 19500
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 2.89,
|
246 |
+
"learning_rate": 0.00031962481962481964,
|
247 |
+
"loss": 0.1908,
|
248 |
+
"step": 20000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.96,
|
252 |
+
"learning_rate": 0.0003151154401154401,
|
253 |
+
"loss": 0.1835,
|
254 |
+
"step": 20500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 3.03,
|
258 |
+
"learning_rate": 0.0003106060606060606,
|
259 |
+
"loss": 0.1755,
|
260 |
+
"step": 21000
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 3.1,
|
264 |
+
"learning_rate": 0.0003060966810966811,
|
265 |
+
"loss": 0.1725,
|
266 |
+
"step": 21500
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 3.17,
|
270 |
+
"learning_rate": 0.0003015873015873016,
|
271 |
+
"loss": 0.172,
|
272 |
+
"step": 22000
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 3.25,
|
276 |
+
"learning_rate": 0.0002970779220779221,
|
277 |
+
"loss": 0.172,
|
278 |
+
"step": 22500
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 3.32,
|
282 |
+
"learning_rate": 0.0002925685425685426,
|
283 |
+
"loss": 0.1649,
|
284 |
+
"step": 23000
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 3.39,
|
288 |
+
"learning_rate": 0.0002880591630591631,
|
289 |
+
"loss": 0.1683,
|
290 |
+
"step": 23500
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 3.46,
|
294 |
+
"learning_rate": 0.00028354978354978356,
|
295 |
+
"loss": 0.1696,
|
296 |
+
"step": 24000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 3.54,
|
300 |
+
"learning_rate": 0.00027904040404040404,
|
301 |
+
"loss": 0.1695,
|
302 |
+
"step": 24500
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 3.61,
|
306 |
+
"learning_rate": 0.0002745310245310245,
|
307 |
+
"loss": 0.1705,
|
308 |
+
"step": 25000
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 3.68,
|
312 |
+
"learning_rate": 0.000270021645021645,
|
313 |
+
"loss": 0.1659,
|
314 |
+
"step": 25500
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 3.75,
|
318 |
+
"learning_rate": 0.0002655122655122655,
|
319 |
+
"loss": 0.1694,
|
320 |
+
"step": 26000
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 3.82,
|
324 |
+
"learning_rate": 0.000261002886002886,
|
325 |
+
"loss": 0.1599,
|
326 |
+
"step": 26500
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 3.9,
|
330 |
+
"learning_rate": 0.0002564935064935065,
|
331 |
+
"loss": 0.1638,
|
332 |
+
"step": 27000
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 3.97,
|
336 |
+
"learning_rate": 0.000251984126984127,
|
337 |
+
"loss": 0.1662,
|
338 |
+
"step": 27500
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 4.04,
|
342 |
+
"learning_rate": 0.0002474747474747475,
|
343 |
+
"loss": 0.1587,
|
344 |
+
"step": 28000
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 4.11,
|
348 |
+
"learning_rate": 0.00024296536796536796,
|
349 |
+
"loss": 0.1453,
|
350 |
+
"step": 28500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 4.18,
|
354 |
+
"learning_rate": 0.00023845598845598847,
|
355 |
+
"loss": 0.1505,
|
356 |
+
"step": 29000
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 4.26,
|
360 |
+
"learning_rate": 0.00023394660894660897,
|
361 |
+
"loss": 0.1498,
|
362 |
+
"step": 29500
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 4.33,
|
366 |
+
"learning_rate": 0.00022943722943722945,
|
367 |
+
"loss": 0.145,
|
368 |
+
"step": 30000
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 4.4,
|
372 |
+
"learning_rate": 0.00022492784992784992,
|
373 |
+
"loss": 0.1446,
|
374 |
+
"step": 30500
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 4.47,
|
378 |
+
"learning_rate": 0.00022041847041847043,
|
379 |
+
"loss": 0.1425,
|
380 |
+
"step": 31000
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 4.55,
|
384 |
+
"learning_rate": 0.0002159090909090909,
|
385 |
+
"loss": 0.1465,
|
386 |
+
"step": 31500
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 4.62,
|
390 |
+
"learning_rate": 0.0002113997113997114,
|
391 |
+
"loss": 0.1501,
|
392 |
+
"step": 32000
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 4.69,
|
396 |
+
"learning_rate": 0.00020689033189033191,
|
397 |
+
"loss": 0.1488,
|
398 |
+
"step": 32500
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 4.76,
|
402 |
+
"learning_rate": 0.0002023809523809524,
|
403 |
+
"loss": 0.1506,
|
404 |
+
"step": 33000
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 4.83,
|
408 |
+
"learning_rate": 0.00019787157287157287,
|
409 |
+
"loss": 0.1457,
|
410 |
+
"step": 33500
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 4.91,
|
414 |
+
"learning_rate": 0.00019336219336219337,
|
415 |
+
"loss": 0.1436,
|
416 |
+
"step": 34000
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 4.98,
|
420 |
+
"learning_rate": 0.00018885281385281385,
|
421 |
+
"loss": 0.1447,
|
422 |
+
"step": 34500
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 5.05,
|
426 |
+
"learning_rate": 0.00018434343434343435,
|
427 |
+
"loss": 0.1381,
|
428 |
+
"step": 35000
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 5.12,
|
432 |
+
"learning_rate": 0.00017983405483405486,
|
433 |
+
"loss": 0.1297,
|
434 |
+
"step": 35500
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 5.19,
|
438 |
+
"learning_rate": 0.00017532467532467534,
|
439 |
+
"loss": 0.1284,
|
440 |
+
"step": 36000
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 5.27,
|
444 |
+
"learning_rate": 0.0001708152958152958,
|
445 |
+
"loss": 0.1375,
|
446 |
+
"step": 36500
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 5.34,
|
450 |
+
"learning_rate": 0.00016630591630591632,
|
451 |
+
"loss": 0.1326,
|
452 |
+
"step": 37000
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 5.41,
|
456 |
+
"learning_rate": 0.0001617965367965368,
|
457 |
+
"loss": 0.1351,
|
458 |
+
"step": 37500
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 5.48,
|
462 |
+
"learning_rate": 0.00015728715728715727,
|
463 |
+
"loss": 0.1338,
|
464 |
+
"step": 38000
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 5.56,
|
468 |
+
"learning_rate": 0.0001527777777777778,
|
469 |
+
"loss": 0.1303,
|
470 |
+
"step": 38500
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 5.63,
|
474 |
+
"learning_rate": 0.00014826839826839828,
|
475 |
+
"loss": 0.1311,
|
476 |
+
"step": 39000
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 5.7,
|
480 |
+
"learning_rate": 0.00014375901875901876,
|
481 |
+
"loss": 0.1332,
|
482 |
+
"step": 39500
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 5.77,
|
486 |
+
"learning_rate": 0.00013924963924963926,
|
487 |
+
"loss": 0.132,
|
488 |
+
"step": 40000
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 5.84,
|
492 |
+
"learning_rate": 0.00013474025974025974,
|
493 |
+
"loss": 0.1299,
|
494 |
+
"step": 40500
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 5.92,
|
498 |
+
"learning_rate": 0.00013023088023088021,
|
499 |
+
"loss": 0.128,
|
500 |
+
"step": 41000
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 5.99,
|
504 |
+
"learning_rate": 0.00012572150072150075,
|
505 |
+
"loss": 0.131,
|
506 |
+
"step": 41500
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 6.06,
|
510 |
+
"learning_rate": 0.00012121212121212122,
|
511 |
+
"loss": 0.1173,
|
512 |
+
"step": 42000
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 6.13,
|
516 |
+
"learning_rate": 0.0001167027417027417,
|
517 |
+
"loss": 0.1195,
|
518 |
+
"step": 42500
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 6.2,
|
522 |
+
"learning_rate": 0.00011219336219336219,
|
523 |
+
"loss": 0.1184,
|
524 |
+
"step": 43000
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 6.28,
|
528 |
+
"learning_rate": 0.0001076839826839827,
|
529 |
+
"loss": 0.1226,
|
530 |
+
"step": 43500
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 6.35,
|
534 |
+
"learning_rate": 0.00010317460317460317,
|
535 |
+
"loss": 0.1182,
|
536 |
+
"step": 44000
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 6.42,
|
540 |
+
"learning_rate": 9.866522366522366e-05,
|
541 |
+
"loss": 0.1249,
|
542 |
+
"step": 44500
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 6.49,
|
546 |
+
"learning_rate": 9.415584415584417e-05,
|
547 |
+
"loss": 0.1187,
|
548 |
+
"step": 45000
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 6.57,
|
552 |
+
"learning_rate": 8.964646464646464e-05,
|
553 |
+
"loss": 0.1214,
|
554 |
+
"step": 45500
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 6.64,
|
558 |
+
"learning_rate": 8.513708513708513e-05,
|
559 |
+
"loss": 0.1194,
|
560 |
+
"step": 46000
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 6.71,
|
564 |
+
"learning_rate": 8.062770562770564e-05,
|
565 |
+
"loss": 0.1187,
|
566 |
+
"step": 46500
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 6.78,
|
570 |
+
"learning_rate": 7.611832611832612e-05,
|
571 |
+
"loss": 0.1168,
|
572 |
+
"step": 47000
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 6.85,
|
576 |
+
"learning_rate": 7.16089466089466e-05,
|
577 |
+
"loss": 0.12,
|
578 |
+
"step": 47500
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 6.93,
|
582 |
+
"learning_rate": 6.709956709956711e-05,
|
583 |
+
"loss": 0.1198,
|
584 |
+
"step": 48000
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 7.0,
|
588 |
+
"learning_rate": 6.259018759018759e-05,
|
589 |
+
"loss": 0.1189,
|
590 |
+
"step": 48500
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 7.07,
|
594 |
+
"learning_rate": 5.808080808080808e-05,
|
595 |
+
"loss": 0.1116,
|
596 |
+
"step": 49000
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 7.14,
|
600 |
+
"learning_rate": 5.357142857142857e-05,
|
601 |
+
"loss": 0.1114,
|
602 |
+
"step": 49500
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 7.21,
|
606 |
+
"learning_rate": 4.9062049062049066e-05,
|
607 |
+
"loss": 0.1108,
|
608 |
+
"step": 50000
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 7.29,
|
612 |
+
"learning_rate": 4.455266955266955e-05,
|
613 |
+
"loss": 0.1101,
|
614 |
+
"step": 50500
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 7.36,
|
618 |
+
"learning_rate": 4.004329004329004e-05,
|
619 |
+
"loss": 0.1062,
|
620 |
+
"step": 51000
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 7.43,
|
624 |
+
"learning_rate": 3.553391053391054e-05,
|
625 |
+
"loss": 0.1141,
|
626 |
+
"step": 51500
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 7.5,
|
630 |
+
"learning_rate": 3.102453102453102e-05,
|
631 |
+
"loss": 0.1106,
|
632 |
+
"step": 52000
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 7.58,
|
636 |
+
"learning_rate": 2.6515151515151516e-05,
|
637 |
+
"loss": 0.1105,
|
638 |
+
"step": 52500
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 7.65,
|
642 |
+
"learning_rate": 2.2005772005772003e-05,
|
643 |
+
"loss": 0.11,
|
644 |
+
"step": 53000
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 7.72,
|
648 |
+
"learning_rate": 1.7496392496392497e-05,
|
649 |
+
"loss": 0.1117,
|
650 |
+
"step": 53500
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 7.79,
|
654 |
+
"learning_rate": 1.2987012987012988e-05,
|
655 |
+
"loss": 0.111,
|
656 |
+
"step": 54000
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 7.86,
|
660 |
+
"learning_rate": 8.477633477633478e-06,
|
661 |
+
"loss": 0.1144,
|
662 |
+
"step": 54500
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 7.94,
|
666 |
+
"learning_rate": 3.968253968253968e-06,
|
667 |
+
"loss": 0.1126,
|
668 |
+
"step": 55000
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 8.0,
|
672 |
+
"step": 55440,
|
673 |
+
"total_flos": 9.377895890180506e+17,
|
674 |
+
"train_loss": 0.18044453823205198,
|
675 |
+
"train_runtime": 140707.6037,
|
676 |
+
"train_samples_per_second": 12.609,
|
677 |
+
"train_steps_per_second": 0.394
|
678 |
+
}
|
679 |
+
],
|
680 |
+
"max_steps": 55440,
|
681 |
+
"num_train_epochs": 8,
|
682 |
+
"total_flos": 9.377895890180506e+17,
|
683 |
+
"trial_name": null,
|
684 |
+
"trial_params": null
|
685 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1119915735c3ff6ede8e35e5249fe528efa032125381474a056d5503ac033301
|
3 |
+
size 3375
|