Upload trainer_state.json
Browse files- trainer_state.json +208 -106
trainer_state.json
CHANGED
@@ -1,202 +1,304 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
"epoch": 1.07,
|
12 |
-
"eval_accuracy": 0.
|
13 |
-
"
|
14 |
-
"
|
15 |
-
"
|
16 |
-
"
|
17 |
-
"
|
|
|
|
|
18 |
"step": 200
|
19 |
},
|
20 |
{
|
21 |
"epoch": 2.14,
|
22 |
-
"eval_accuracy": 0.
|
23 |
-
"
|
24 |
-
"
|
25 |
-
"
|
26 |
-
"
|
27 |
-
"
|
|
|
|
|
28 |
"step": 400
|
29 |
},
|
30 |
{
|
31 |
"epoch": 2.67,
|
32 |
-
"learning_rate": 4.
|
33 |
-
"loss": 1.
|
34 |
"step": 500
|
35 |
},
|
36 |
{
|
37 |
"epoch": 3.21,
|
38 |
-
"eval_accuracy": 0.
|
39 |
-
"
|
40 |
-
"
|
41 |
-
"
|
42 |
-
"
|
43 |
-
"
|
|
|
|
|
44 |
"step": 600
|
45 |
},
|
46 |
{
|
47 |
"epoch": 4.28,
|
48 |
-
"eval_accuracy": 0.
|
49 |
-
"
|
50 |
-
"
|
51 |
-
"
|
52 |
-
"
|
53 |
-
"
|
|
|
|
|
54 |
"step": 800
|
55 |
},
|
56 |
{
|
57 |
"epoch": 5.35,
|
58 |
-
"learning_rate": 4.
|
59 |
-
"loss": 0.
|
60 |
"step": 1000
|
61 |
},
|
62 |
{
|
63 |
"epoch": 5.35,
|
64 |
-
"eval_accuracy": 0.
|
65 |
-
"
|
66 |
-
"
|
67 |
-
"
|
68 |
-
"
|
69 |
-
"
|
|
|
|
|
70 |
"step": 1000
|
71 |
},
|
72 |
{
|
73 |
"epoch": 6.42,
|
74 |
-
"eval_accuracy": 0.
|
75 |
-
"
|
76 |
-
"
|
77 |
-
"
|
78 |
-
"
|
79 |
-
"
|
|
|
|
|
80 |
"step": 1200
|
81 |
},
|
82 |
{
|
83 |
"epoch": 7.49,
|
84 |
-
"eval_accuracy": 0.
|
85 |
-
"
|
86 |
-
"
|
87 |
-
"
|
88 |
-
"
|
89 |
-
"
|
|
|
|
|
90 |
"step": 1400
|
91 |
},
|
92 |
{
|
93 |
"epoch": 8.02,
|
94 |
-
"learning_rate":
|
95 |
-
"loss": 0.
|
96 |
"step": 1500
|
97 |
},
|
98 |
{
|
99 |
"epoch": 8.56,
|
100 |
-
"eval_accuracy": 0.
|
101 |
-
"
|
102 |
-
"
|
103 |
-
"
|
104 |
-
"
|
105 |
-
"
|
|
|
|
|
106 |
"step": 1600
|
107 |
},
|
108 |
{
|
109 |
"epoch": 9.63,
|
110 |
-
"eval_accuracy": 0.
|
111 |
-
"
|
112 |
-
"
|
113 |
-
"
|
114 |
-
"
|
115 |
-
"
|
|
|
|
|
116 |
"step": 1800
|
117 |
},
|
118 |
{
|
119 |
"epoch": 10.7,
|
120 |
-
"learning_rate": 3.
|
121 |
-
"loss": 0.
|
122 |
"step": 2000
|
123 |
},
|
124 |
{
|
125 |
"epoch": 10.7,
|
126 |
-
"eval_accuracy": 0.
|
127 |
-
"
|
128 |
-
"
|
129 |
-
"
|
130 |
-
"
|
131 |
-
"
|
|
|
|
|
132 |
"step": 2000
|
133 |
},
|
134 |
{
|
135 |
"epoch": 11.76,
|
136 |
-
"eval_accuracy": 0.
|
137 |
-
"
|
138 |
-
"
|
139 |
-
"
|
140 |
-
"
|
141 |
-
"
|
|
|
|
|
142 |
"step": 2200
|
143 |
},
|
144 |
{
|
145 |
"epoch": 12.83,
|
146 |
-
"eval_accuracy": 0.
|
147 |
-
"
|
148 |
-
"
|
149 |
-
"
|
150 |
-
"
|
151 |
-
"
|
|
|
|
|
152 |
"step": 2400
|
153 |
},
|
154 |
{
|
155 |
"epoch": 13.37,
|
156 |
-
"learning_rate":
|
157 |
-
"loss": 0.
|
158 |
"step": 2500
|
159 |
},
|
160 |
{
|
161 |
"epoch": 13.9,
|
162 |
-
"eval_accuracy": 0.
|
163 |
-
"
|
164 |
-
"
|
165 |
-
"
|
166 |
-
"
|
167 |
-
"
|
|
|
|
|
168 |
"step": 2600
|
169 |
},
|
170 |
{
|
171 |
"epoch": 14.97,
|
172 |
-
"eval_accuracy": 0.
|
173 |
-
"
|
174 |
-
"
|
175 |
-
"
|
176 |
-
"
|
177 |
-
"
|
|
|
|
|
178 |
"step": 2800
|
179 |
},
|
180 |
{
|
181 |
"epoch": 16.04,
|
182 |
-
"learning_rate":
|
183 |
-
"loss": 0.
|
184 |
"step": 3000
|
185 |
},
|
186 |
{
|
187 |
"epoch": 16.04,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
188 |
"eval_accuracy": 0.6898395721925134,
|
189 |
-
"
|
|
|
|
|
190 |
"eval_recall": 0.6898395721925134,
|
191 |
-
"eval_runtime": 10.
|
192 |
-
"eval_samples_per_second": 18.
|
193 |
-
"eval_steps_per_second": 4.
|
194 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
195 |
}
|
196 |
],
|
197 |
-
"max_steps":
|
198 |
-
"num_train_epochs":
|
199 |
-
"total_flos":
|
200 |
"trial_name": null,
|
201 |
"trial_params": null
|
202 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 21.390374331550802,
|
5 |
+
"global_step": 4000,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
"epoch": 1.07,
|
12 |
+
"eval_accuracy": 0.48128342245989303,
|
13 |
+
"eval_f1": 0.48128342245989303,
|
14 |
+
"eval_loss": 1.350968837738037,
|
15 |
+
"eval_precision": 0.48128342245989303,
|
16 |
+
"eval_recall": 0.48128342245989303,
|
17 |
+
"eval_runtime": 10.797,
|
18 |
+
"eval_samples_per_second": 17.32,
|
19 |
+
"eval_steps_per_second": 4.353,
|
20 |
"step": 200
|
21 |
},
|
22 |
{
|
23 |
"epoch": 2.14,
|
24 |
+
"eval_accuracy": 0.6577540106951871,
|
25 |
+
"eval_f1": 0.6577540106951871,
|
26 |
+
"eval_loss": 1.0548419952392578,
|
27 |
+
"eval_precision": 0.6577540106951871,
|
28 |
+
"eval_recall": 0.6577540106951871,
|
29 |
+
"eval_runtime": 10.384,
|
30 |
+
"eval_samples_per_second": 18.008,
|
31 |
+
"eval_steps_per_second": 4.526,
|
32 |
"step": 400
|
33 |
},
|
34 |
{
|
35 |
"epoch": 2.67,
|
36 |
+
"learning_rate": 4.732620320855615e-05,
|
37 |
+
"loss": 1.271,
|
38 |
"step": 500
|
39 |
},
|
40 |
{
|
41 |
"epoch": 3.21,
|
42 |
+
"eval_accuracy": 0.6631016042780749,
|
43 |
+
"eval_f1": 0.6631016042780749,
|
44 |
+
"eval_loss": 1.257256031036377,
|
45 |
+
"eval_precision": 0.6631016042780749,
|
46 |
+
"eval_recall": 0.6631016042780749,
|
47 |
+
"eval_runtime": 10.09,
|
48 |
+
"eval_samples_per_second": 18.533,
|
49 |
+
"eval_steps_per_second": 4.658,
|
50 |
"step": 600
|
51 |
},
|
52 |
{
|
53 |
"epoch": 4.28,
|
54 |
+
"eval_accuracy": 0.6577540106951871,
|
55 |
+
"eval_f1": 0.6577540106951871,
|
56 |
+
"eval_loss": 1.3473643064498901,
|
57 |
+
"eval_precision": 0.6577540106951871,
|
58 |
+
"eval_recall": 0.6577540106951871,
|
59 |
+
"eval_runtime": 10.111,
|
60 |
+
"eval_samples_per_second": 18.495,
|
61 |
+
"eval_steps_per_second": 4.648,
|
62 |
"step": 800
|
63 |
},
|
64 |
{
|
65 |
"epoch": 5.35,
|
66 |
+
"learning_rate": 4.4652406417112304e-05,
|
67 |
+
"loss": 0.791,
|
68 |
"step": 1000
|
69 |
},
|
70 |
{
|
71 |
"epoch": 5.35,
|
72 |
+
"eval_accuracy": 0.6844919786096256,
|
73 |
+
"eval_f1": 0.6844919786096256,
|
74 |
+
"eval_loss": 1.6341558694839478,
|
75 |
+
"eval_precision": 0.6844919786096256,
|
76 |
+
"eval_recall": 0.6844919786096256,
|
77 |
+
"eval_runtime": 10.125,
|
78 |
+
"eval_samples_per_second": 18.469,
|
79 |
+
"eval_steps_per_second": 4.642,
|
80 |
"step": 1000
|
81 |
},
|
82 |
{
|
83 |
"epoch": 6.42,
|
84 |
+
"eval_accuracy": 0.7272727272727273,
|
85 |
+
"eval_f1": 0.7272727272727273,
|
86 |
+
"eval_loss": 1.7296054363250732,
|
87 |
+
"eval_precision": 0.7272727272727273,
|
88 |
+
"eval_recall": 0.7272727272727273,
|
89 |
+
"eval_runtime": 10.343,
|
90 |
+
"eval_samples_per_second": 18.08,
|
91 |
+
"eval_steps_per_second": 4.544,
|
92 |
"step": 1200
|
93 |
},
|
94 |
{
|
95 |
"epoch": 7.49,
|
96 |
+
"eval_accuracy": 0.7112299465240641,
|
97 |
+
"eval_f1": 0.7112299465240641,
|
98 |
+
"eval_loss": 1.687750220298767,
|
99 |
+
"eval_precision": 0.7112299465240641,
|
100 |
+
"eval_recall": 0.7112299465240641,
|
101 |
+
"eval_runtime": 10.345,
|
102 |
+
"eval_samples_per_second": 18.076,
|
103 |
+
"eval_steps_per_second": 4.543,
|
104 |
"step": 1400
|
105 |
},
|
106 |
{
|
107 |
"epoch": 8.02,
|
108 |
+
"learning_rate": 4.197860962566845e-05,
|
109 |
+
"loss": 0.5343,
|
110 |
"step": 1500
|
111 |
},
|
112 |
{
|
113 |
"epoch": 8.56,
|
114 |
+
"eval_accuracy": 0.6951871657754011,
|
115 |
+
"eval_f1": 0.6951871657754011,
|
116 |
+
"eval_loss": 1.8570679426193237,
|
117 |
+
"eval_precision": 0.6951871657754011,
|
118 |
+
"eval_recall": 0.6951871657754011,
|
119 |
+
"eval_runtime": 10.19,
|
120 |
+
"eval_samples_per_second": 18.351,
|
121 |
+
"eval_steps_per_second": 4.612,
|
122 |
"step": 1600
|
123 |
},
|
124 |
{
|
125 |
"epoch": 9.63,
|
126 |
+
"eval_accuracy": 0.7058823529411765,
|
127 |
+
"eval_f1": 0.7058823529411765,
|
128 |
+
"eval_loss": 1.9290939569473267,
|
129 |
+
"eval_precision": 0.7058823529411765,
|
130 |
+
"eval_recall": 0.7058823529411765,
|
131 |
+
"eval_runtime": 10.161,
|
132 |
+
"eval_samples_per_second": 18.404,
|
133 |
+
"eval_steps_per_second": 4.626,
|
134 |
"step": 1800
|
135 |
},
|
136 |
{
|
137 |
"epoch": 10.7,
|
138 |
+
"learning_rate": 3.93048128342246e-05,
|
139 |
+
"loss": 0.2934,
|
140 |
"step": 2000
|
141 |
},
|
142 |
{
|
143 |
"epoch": 10.7,
|
144 |
+
"eval_accuracy": 0.7005347593582888,
|
145 |
+
"eval_f1": 0.7005347593582888,
|
146 |
+
"eval_loss": 1.9618135690689087,
|
147 |
+
"eval_precision": 0.7005347593582888,
|
148 |
+
"eval_recall": 0.7005347593582888,
|
149 |
+
"eval_runtime": 10.233,
|
150 |
+
"eval_samples_per_second": 18.274,
|
151 |
+
"eval_steps_per_second": 4.593,
|
152 |
"step": 2000
|
153 |
},
|
154 |
{
|
155 |
"epoch": 11.76,
|
156 |
+
"eval_accuracy": 0.7272727272727273,
|
157 |
+
"eval_f1": 0.7272727272727273,
|
158 |
+
"eval_loss": 1.8590352535247803,
|
159 |
+
"eval_precision": 0.7272727272727273,
|
160 |
+
"eval_recall": 0.7272727272727273,
|
161 |
+
"eval_runtime": 10.155,
|
162 |
+
"eval_samples_per_second": 18.415,
|
163 |
+
"eval_steps_per_second": 4.628,
|
164 |
"step": 2200
|
165 |
},
|
166 |
{
|
167 |
"epoch": 12.83,
|
168 |
+
"eval_accuracy": 0.732620320855615,
|
169 |
+
"eval_f1": 0.7326203208556151,
|
170 |
+
"eval_loss": 2.042591094970703,
|
171 |
+
"eval_precision": 0.732620320855615,
|
172 |
+
"eval_recall": 0.732620320855615,
|
173 |
+
"eval_runtime": 10.107,
|
174 |
+
"eval_samples_per_second": 18.502,
|
175 |
+
"eval_steps_per_second": 4.65,
|
176 |
"step": 2400
|
177 |
},
|
178 |
{
|
179 |
"epoch": 13.37,
|
180 |
+
"learning_rate": 3.6631016042780753e-05,
|
181 |
+
"loss": 0.1892,
|
182 |
"step": 2500
|
183 |
},
|
184 |
{
|
185 |
"epoch": 13.9,
|
186 |
+
"eval_accuracy": 0.7058823529411765,
|
187 |
+
"eval_f1": 0.7058823529411765,
|
188 |
+
"eval_loss": 2.264678716659546,
|
189 |
+
"eval_precision": 0.7058823529411765,
|
190 |
+
"eval_recall": 0.7058823529411765,
|
191 |
+
"eval_runtime": 10.161,
|
192 |
+
"eval_samples_per_second": 18.404,
|
193 |
+
"eval_steps_per_second": 4.626,
|
194 |
"step": 2600
|
195 |
},
|
196 |
{
|
197 |
"epoch": 14.97,
|
198 |
+
"eval_accuracy": 0.7112299465240641,
|
199 |
+
"eval_f1": 0.7112299465240641,
|
200 |
+
"eval_loss": 2.111647844314575,
|
201 |
+
"eval_precision": 0.7112299465240641,
|
202 |
+
"eval_recall": 0.7112299465240641,
|
203 |
+
"eval_runtime": 10.145,
|
204 |
+
"eval_samples_per_second": 18.433,
|
205 |
+
"eval_steps_per_second": 4.633,
|
206 |
"step": 2800
|
207 |
},
|
208 |
{
|
209 |
"epoch": 16.04,
|
210 |
+
"learning_rate": 3.39572192513369e-05,
|
211 |
+
"loss": 0.129,
|
212 |
"step": 3000
|
213 |
},
|
214 |
{
|
215 |
"epoch": 16.04,
|
216 |
+
"eval_accuracy": 0.7165775401069518,
|
217 |
+
"eval_f1": 0.7165775401069518,
|
218 |
+
"eval_loss": 2.1683263778686523,
|
219 |
+
"eval_precision": 0.7165775401069518,
|
220 |
+
"eval_recall": 0.7165775401069518,
|
221 |
+
"eval_runtime": 10.287,
|
222 |
+
"eval_samples_per_second": 18.178,
|
223 |
+
"eval_steps_per_second": 4.569,
|
224 |
+
"step": 3000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 17.11,
|
228 |
+
"eval_accuracy": 0.7165775401069518,
|
229 |
+
"eval_f1": 0.7165775401069518,
|
230 |
+
"eval_loss": 2.329145669937134,
|
231 |
+
"eval_precision": 0.7165775401069518,
|
232 |
+
"eval_recall": 0.7165775401069518,
|
233 |
+
"eval_runtime": 10.188,
|
234 |
+
"eval_samples_per_second": 18.355,
|
235 |
+
"eval_steps_per_second": 4.613,
|
236 |
+
"step": 3200
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 18.18,
|
240 |
"eval_accuracy": 0.6898395721925134,
|
241 |
+
"eval_f1": 0.6898395721925134,
|
242 |
+
"eval_loss": 2.538438081741333,
|
243 |
+
"eval_precision": 0.6898395721925134,
|
244 |
"eval_recall": 0.6898395721925134,
|
245 |
+
"eval_runtime": 10.202,
|
246 |
+
"eval_samples_per_second": 18.33,
|
247 |
+
"eval_steps_per_second": 4.607,
|
248 |
+
"step": 3400
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 18.72,
|
252 |
+
"learning_rate": 3.128342245989305e-05,
|
253 |
+
"loss": 0.1084,
|
254 |
+
"step": 3500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 19.25,
|
258 |
+
"eval_accuracy": 0.6951871657754011,
|
259 |
+
"eval_f1": 0.6951871657754011,
|
260 |
+
"eval_loss": 2.630084753036499,
|
261 |
+
"eval_precision": 0.6951871657754011,
|
262 |
+
"eval_recall": 0.6951871657754011,
|
263 |
+
"eval_runtime": 10.362,
|
264 |
+
"eval_samples_per_second": 18.047,
|
265 |
+
"eval_steps_per_second": 4.536,
|
266 |
+
"step": 3600
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 20.32,
|
270 |
+
"eval_accuracy": 0.6684491978609626,
|
271 |
+
"eval_f1": 0.6684491978609626,
|
272 |
+
"eval_loss": 2.8742926120758057,
|
273 |
+
"eval_precision": 0.6684491978609626,
|
274 |
+
"eval_recall": 0.6684491978609626,
|
275 |
+
"eval_runtime": 10.323,
|
276 |
+
"eval_samples_per_second": 18.115,
|
277 |
+
"eval_steps_per_second": 4.553,
|
278 |
+
"step": 3800
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 21.39,
|
282 |
+
"learning_rate": 2.8609625668449196e-05,
|
283 |
+
"loss": 0.0935,
|
284 |
+
"step": 4000
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 21.39,
|
288 |
+
"eval_accuracy": 0.7219251336898396,
|
289 |
+
"eval_f1": 0.7219251336898396,
|
290 |
+
"eval_loss": 2.320441484451294,
|
291 |
+
"eval_precision": 0.7219251336898396,
|
292 |
+
"eval_recall": 0.7219251336898396,
|
293 |
+
"eval_runtime": 10.306,
|
294 |
+
"eval_samples_per_second": 18.145,
|
295 |
+
"eval_steps_per_second": 4.56,
|
296 |
+
"step": 4000
|
297 |
}
|
298 |
],
|
299 |
+
"max_steps": 9350,
|
300 |
+
"num_train_epochs": 50,
|
301 |
+
"total_flos": 4198877016109056.0,
|
302 |
"trial_name": null,
|
303 |
"trial_params": null
|
304 |
}
|