Ubuntu commited on
Commit
1fda2ee
1 Parent(s): 680526d

update qqp

Browse files
all_results.json CHANGED
@@ -1,16 +1,15 @@
1
  {
2
  "epoch": 10.0,
3
- "eval_accuracy": 0.6318327974276527,
4
- "eval_combined_score": 0.3159163987138264,
5
- "eval_f1": 0.0,
6
- "eval_loss": 1.2188094854354858,
7
- "eval_runtime": 186.2619,
8
  "eval_samples": 40430,
9
- "eval_samples_per_second": 217.06,
10
- "eval_steps_per_second": 3.393,
11
- "train_loss": 0.627387775952303,
12
- "train_runtime": 15793.4855,
13
  "train_samples": 363846,
14
- "train_samples_per_second": 230.377,
15
- "train_steps_per_second": 1.2
16
- }
1
  {
2
  "epoch": 10.0,
3
+ "eval_accuracy": 0.9279495424189957,
4
+ "eval_combined_score": 0.9155458435286998,
5
+ "eval_f1": 0.903142144638404,
6
+ "eval_loss": 0.546863317489624,
7
+ "eval_runtime": 230.8888,
8
  "eval_samples": 40430,
9
+ "eval_samples_per_second": 175.106,
10
+ "eval_steps_per_second": 5.474,
11
+ "train_runtime": 69350.928,
 
12
  "train_samples": 363846,
13
+ "train_samples_per_second": 52.464,
14
+ "train_steps_per_second": 0.82
15
+ }
eval_results.json CHANGED
@@ -1,11 +1,11 @@
1
  {
2
  "epoch": 10.0,
3
- "eval_accuracy": 0.6318327974276527,
4
- "eval_combined_score": 0.3159163987138264,
5
- "eval_f1": 0.0,
6
- "eval_loss": 1.2188094854354858,
7
- "eval_runtime": 186.2619,
8
  "eval_samples": 40430,
9
- "eval_samples_per_second": 217.06,
10
- "eval_steps_per_second": 3.393
11
  }
1
  {
2
  "epoch": 10.0,
3
+ "eval_accuracy": 0.9279495424189957,
4
+ "eval_combined_score": 0.9155458435286998,
5
+ "eval_f1": 0.903142144638404,
6
+ "eval_loss": 0.546863317489624,
7
+ "eval_runtime": 230.8888,
8
  "eval_samples": 40430,
9
+ "eval_samples_per_second": 175.106,
10
+ "eval_steps_per_second": 5.474
11
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2bad3756bb3c8138ba00e840604a1159a3f116838d94a1aa51b70c58364ad4bb
3
- size 1340745306
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9d16d2340b641366c4006a5c2d7277313853154b2c6c6ab0019a21b97fc1524
3
+ size 1340746075
train_results.json CHANGED
@@ -1,8 +1,7 @@
1
  {
2
  "epoch": 10.0,
3
- "train_loss": 0.627387775952303,
4
- "train_runtime": 15793.4855,
5
  "train_samples": 363846,
6
- "train_samples_per_second": 230.377,
7
- "train_steps_per_second": 1.2
8
  }
1
  {
2
  "epoch": 10.0,
3
+ "train_runtime": 69350.928,
 
4
  "train_samples": 363846,
5
+ "train_samples_per_second": 52.464,
6
+ "train_steps_per_second": 0.82
7
  }
trainer_state.json CHANGED
@@ -2,246 +2,701 @@
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
  "epoch": 10.0,
5
- "global_step": 18960,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
- "epoch": 0.26,
12
- "learning_rate": 1.947257383966245e-05,
13
- "loss": 0.3021,
14
  "step": 500
15
  },
16
  {
17
- "epoch": 0.53,
18
- "learning_rate": 1.8945147679324897e-05,
19
- "loss": 0.5336,
20
  "step": 1000
21
  },
22
  {
23
- "epoch": 0.79,
24
- "learning_rate": 1.8417721518987345e-05,
25
- "loss": 0.6486,
26
  "step": 1500
27
  },
28
  {
29
- "epoch": 1.05,
30
- "learning_rate": 1.789029535864979e-05,
31
- "loss": 0.6483,
32
  "step": 2000
33
  },
34
  {
35
- "epoch": 1.32,
36
- "learning_rate": 1.7362869198312237e-05,
37
- "loss": 0.6448,
38
  "step": 2500
39
  },
40
  {
41
- "epoch": 1.58,
42
- "learning_rate": 1.6835443037974685e-05,
43
- "loss": 0.6453,
44
  "step": 3000
45
  },
46
  {
47
- "epoch": 1.85,
48
- "learning_rate": 1.6308016877637133e-05,
49
- "loss": 0.6458,
50
  "step": 3500
51
  },
52
  {
53
- "epoch": 2.11,
54
- "learning_rate": 1.578059071729958e-05,
55
- "loss": 0.6441,
56
  "step": 4000
57
  },
58
  {
59
- "epoch": 2.37,
60
- "learning_rate": 1.5253164556962025e-05,
61
- "loss": 0.6428,
62
  "step": 4500
63
  },
64
  {
65
- "epoch": 2.64,
66
- "learning_rate": 1.4725738396624474e-05,
67
- "loss": 0.6428,
68
  "step": 5000
69
  },
70
  {
71
- "epoch": 2.9,
72
- "learning_rate": 1.4198312236286922e-05,
73
- "loss": 0.6404,
74
  "step": 5500
75
  },
76
  {
77
- "epoch": 3.16,
78
- "learning_rate": 1.3670886075949368e-05,
79
- "loss": 0.6399,
80
  "step": 6000
81
  },
82
  {
83
- "epoch": 3.43,
84
- "learning_rate": 1.3143459915611816e-05,
85
- "loss": 0.6411,
86
  "step": 6500
87
  },
88
  {
89
- "epoch": 3.69,
90
- "learning_rate": 1.2616033755274262e-05,
91
- "loss": 0.6393,
92
  "step": 7000
93
  },
94
  {
95
- "epoch": 3.96,
96
- "learning_rate": 1.208860759493671e-05,
97
- "loss": 0.6398,
98
  "step": 7500
99
  },
100
  {
101
- "epoch": 4.22,
102
- "learning_rate": 1.1561181434599158e-05,
103
- "loss": 0.6405,
104
  "step": 8000
105
  },
106
  {
107
- "epoch": 4.48,
108
- "learning_rate": 1.1033755274261604e-05,
109
- "loss": 0.6389,
110
  "step": 8500
111
  },
112
  {
113
- "epoch": 4.75,
114
- "learning_rate": 1.0506329113924052e-05,
115
- "loss": 0.6367,
116
  "step": 9000
117
  },
118
  {
119
- "epoch": 5.01,
120
- "learning_rate": 9.9789029535865e-06,
121
- "loss": 0.6379,
122
  "step": 9500
123
  },
124
  {
125
- "epoch": 5.27,
126
- "learning_rate": 9.451476793248946e-06,
127
- "loss": 0.638,
128
  "step": 10000
129
  },
130
  {
131
- "epoch": 5.54,
132
- "learning_rate": 8.924050632911393e-06,
133
- "loss": 0.6382,
134
  "step": 10500
135
  },
136
  {
137
- "epoch": 5.8,
138
- "learning_rate": 8.39662447257384e-06,
139
- "loss": 0.637,
140
  "step": 11000
141
  },
142
  {
143
- "epoch": 6.07,
144
- "learning_rate": 7.869198312236287e-06,
145
- "loss": 0.6384,
146
  "step": 11500
147
  },
148
  {
149
- "epoch": 6.33,
150
- "learning_rate": 7.341772151898735e-06,
151
- "loss": 0.637,
152
  "step": 12000
153
  },
154
  {
155
- "epoch": 6.59,
156
- "learning_rate": 6.814345991561182e-06,
157
- "loss": 0.6358,
158
  "step": 12500
159
  },
160
  {
161
- "epoch": 6.86,
162
- "learning_rate": 6.286919831223629e-06,
163
- "loss": 0.6352,
164
  "step": 13000
165
  },
166
  {
167
- "epoch": 7.12,
168
- "learning_rate": 5.759493670886076e-06,
169
- "loss": 0.6375,
170
  "step": 13500
171
  },
172
  {
173
- "epoch": 7.38,
174
- "learning_rate": 5.2320675105485245e-06,
175
- "loss": 0.6356,
176
  "step": 14000
177
  },
178
  {
179
- "epoch": 7.65,
180
- "learning_rate": 4.7046413502109714e-06,
181
- "loss": 0.6375,
182
  "step": 14500
183
  },
184
  {
185
- "epoch": 7.91,
186
- "learning_rate": 4.177215189873418e-06,
187
- "loss": 0.6358,
188
  "step": 15000
189
  },
190
  {
191
- "epoch": 8.18,
192
- "learning_rate": 3.649789029535865e-06,
193
- "loss": 0.6378,
194
  "step": 15500
195
  },
196
  {
197
- "epoch": 8.44,
198
- "learning_rate": 3.1223628691983127e-06,
199
- "loss": 0.6343,
200
  "step": 16000
201
  },
202
  {
203
- "epoch": 8.7,
204
- "learning_rate": 2.5949367088607596e-06,
205
- "loss": 0.6349,
206
  "step": 16500
207
  },
208
  {
209
- "epoch": 8.97,
210
- "learning_rate": 2.067510548523207e-06,
211
- "loss": 0.6363,
212
  "step": 17000
213
  },
214
  {
215
- "epoch": 9.23,
216
- "learning_rate": 1.5400843881856542e-06,
217
- "loss": 0.6343,
218
  "step": 17500
219
  },
220
  {
221
- "epoch": 9.49,
222
- "learning_rate": 1.0126582278481013e-06,
223
- "loss": 0.6345,
224
  "step": 18000
225
  },
226
  {
227
- "epoch": 9.76,
228
- "learning_rate": 4.852320675105486e-07,
229
- "loss": 0.6363,
230
  "step": 18500
231
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
232
  {
233
  "epoch": 10.0,
234
- "step": 18960,
235
- "total_flos": 9.365052001194086e+17,
236
- "train_loss": 0.627387775952303,
237
- "train_runtime": 15793.4855,
238
- "train_samples_per_second": 230.377,
239
- "train_steps_per_second": 1.2
240
  }
241
  ],
242
- "max_steps": 18960,
243
  "num_train_epochs": 10,
244
- "total_flos": 9.365052001194086e+17,
245
  "trial_name": null,
246
  "trial_params": null
247
  }
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
  "epoch": 10.0,
5
+ "global_step": 56860,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
+ "epoch": 0.09,
12
+ "learning_rate": 1.9824129440731624e-05,
13
+ "loss": 0.5315,
14
  "step": 500
15
  },
16
  {
17
+ "epoch": 0.18,
18
+ "learning_rate": 1.9648258881463246e-05,
19
+ "loss": 0.2968,
20
  "step": 1000
21
  },
22
  {
23
+ "epoch": 0.26,
24
+ "learning_rate": 1.9472388322194867e-05,
25
+ "loss": 0.2756,
26
  "step": 1500
27
  },
28
  {
29
+ "epoch": 0.35,
30
+ "learning_rate": 1.929651776292649e-05,
31
+ "loss": 0.2673,
32
  "step": 2000
33
  },
34
  {
35
+ "epoch": 0.44,
36
+ "learning_rate": 1.9120647203658108e-05,
37
+ "loss": 0.2521,
38
  "step": 2500
39
  },
40
  {
41
+ "epoch": 0.53,
42
+ "learning_rate": 1.894477664438973e-05,
43
+ "loss": 0.2405,
44
  "step": 3000
45
  },
46
  {
47
+ "epoch": 0.62,
48
+ "learning_rate": 1.8768906085121352e-05,
49
+ "loss": 0.2264,
50
  "step": 3500
51
  },
52
  {
53
+ "epoch": 0.7,
54
+ "learning_rate": 1.8593035525852974e-05,
55
+ "loss": 0.2264,
56
  "step": 4000
57
  },
58
  {
59
+ "epoch": 0.79,
60
+ "learning_rate": 1.8417164966584596e-05,
61
+ "loss": 0.2231,
62
  "step": 4500
63
  },
64
  {
65
+ "epoch": 0.88,
66
+ "learning_rate": 1.8241294407316218e-05,
67
+ "loss": 0.6134,
68
  "step": 5000
69
  },
70
  {
71
+ "epoch": 0.97,
72
+ "learning_rate": 1.806542384804784e-05,
73
+ "loss": 0.6581,
74
  "step": 5500
75
  },
76
  {
77
+ "epoch": 1.06,
78
+ "learning_rate": 1.788955328877946e-05,
79
+ "loss": 0.6592,
80
  "step": 6000
81
  },
82
  {
83
+ "epoch": 1.14,
84
+ "learning_rate": 1.771368272951108e-05,
85
+ "loss": 0.6561,
86
  "step": 6500
87
  },
88
  {
89
+ "epoch": 1.23,
90
+ "learning_rate": 1.7537812170242702e-05,
91
+ "loss": 0.6599,
92
  "step": 7000
93
  },
94
  {
95
+ "epoch": 1.32,
96
+ "learning_rate": 1.7361941610974324e-05,
97
+ "loss": 0.6569,
98
  "step": 7500
99
  },
100
  {
101
+ "epoch": 1.41,
102
+ "learning_rate": 1.7186071051705946e-05,
103
+ "loss": 0.6531,
104
  "step": 8000
105
  },
106
  {
107
+ "epoch": 1.49,
108
+ "learning_rate": 1.7010200492437568e-05,
109
+ "loss": 0.2212,
110
  "step": 8500
111
  },
112
  {
113
+ "epoch": 1.58,
114
+ "learning_rate": 1.683432993316919e-05,
115
+ "loss": 0.4708,
116
  "step": 9000
117
  },
118
  {
119
+ "epoch": 1.67,
120
+ "learning_rate": 1.665845937390081e-05,
121
+ "loss": 0.2605,
122
  "step": 9500
123
  },
124
  {
125
+ "epoch": 1.76,
126
+ "learning_rate": 1.6482588814632434e-05,
127
+ "loss": 0.1987,
128
  "step": 10000
129
  },
130
  {
131
+ "epoch": 1.85,
132
+ "learning_rate": 1.6306718255364052e-05,
133
+ "loss": 0.1908,
134
  "step": 10500
135
  },
136
  {
137
+ "epoch": 1.93,
138
+ "learning_rate": 1.6130847696095674e-05,
139
+ "loss": 0.185,
140
  "step": 11000
141
  },
142
  {
143
+ "epoch": 2.02,
144
+ "learning_rate": 1.5954977136827296e-05,
145
+ "loss": 0.1767,
146
  "step": 11500
147
  },
148
  {
149
+ "epoch": 2.11,
150
+ "learning_rate": 1.5779106577558918e-05,
151
+ "loss": 0.1551,
152
  "step": 12000
153
  },
154
  {
155
+ "epoch": 2.2,
156
+ "learning_rate": 1.560323601829054e-05,
157
+ "loss": 0.1476,
158
  "step": 12500
159
  },
160
  {
161
+ "epoch": 2.29,
162
+ "learning_rate": 1.5427365459022162e-05,
163
+ "loss": 0.1449,
164
  "step": 13000
165
  },
166
  {
167
+ "epoch": 2.37,
168
+ "learning_rate": 1.5251494899753782e-05,
169
+ "loss": 0.1566,
170
  "step": 13500
171
  },
172
  {
173
+ "epoch": 2.46,
174
+ "learning_rate": 1.5075624340485404e-05,
175
+ "loss": 0.1586,
176
  "step": 14000
177
  },
178
  {
179
+ "epoch": 2.55,
180
+ "learning_rate": 1.4899753781217026e-05,
181
+ "loss": 0.1487,
182
  "step": 14500
183
  },
184
  {
185
+ "epoch": 2.64,
186
+ "learning_rate": 1.4723883221948648e-05,
187
+ "loss": 0.1493,
188
  "step": 15000
189
  },
190
  {
191
+ "epoch": 2.73,
192
+ "learning_rate": 1.4548012662680268e-05,
193
+ "loss": 0.1547,
194
  "step": 15500
195
  },
196
  {
197
+ "epoch": 2.81,
198
+ "learning_rate": 1.437214210341189e-05,
199
+ "loss": 0.1528,
200
  "step": 16000
201
  },
202
  {
203
+ "epoch": 2.9,
204
+ "learning_rate": 1.4196271544143512e-05,
205
+ "loss": 0.1447,
206
  "step": 16500
207
  },
208
  {
209
+ "epoch": 2.99,
210
+ "learning_rate": 1.4020400984875134e-05,
211
+ "loss": 0.1562,
212
  "step": 17000
213
  },
214
  {
215
+ "epoch": 3.08,
216
+ "learning_rate": 1.3844530425606754e-05,
217
+ "loss": 0.1022,
218
  "step": 17500
219
  },
220
  {
221
+ "epoch": 3.17,
222
+ "learning_rate": 1.3668659866338376e-05,
223
+ "loss": 0.099,
224
  "step": 18000
225
  },
226
  {
227
+ "epoch": 3.25,
228
+ "learning_rate": 1.3492789307069998e-05,
229
+ "loss": 0.0971,
230
  "step": 18500
231
  },
232
+ {
233
+ "epoch": 3.34,
234
+ "learning_rate": 1.331691874780162e-05,
235
+ "loss": 0.1046,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 3.43,
240
+ "learning_rate": 1.314104818853324e-05,
241
+ "loss": 0.1083,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 3.52,
246
+ "learning_rate": 1.2965177629264862e-05,
247
+ "loss": 0.0998,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 3.61,
252
+ "learning_rate": 1.2789307069996484e-05,
253
+ "loss": 0.1049,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 3.69,
258
+ "learning_rate": 1.2613436510728106e-05,
259
+ "loss": 0.1052,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 3.78,
264
+ "learning_rate": 1.2437565951459726e-05,
265
+ "loss": 0.108,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 3.87,
270
+ "learning_rate": 1.2261695392191348e-05,
271
+ "loss": 0.0991,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 3.96,
276
+ "learning_rate": 1.208582483292297e-05,
277
+ "loss": 0.1024,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 4.05,
282
+ "learning_rate": 1.1909954273654592e-05,
283
+ "loss": 0.0822,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 4.13,
288
+ "learning_rate": 1.1734083714386212e-05,
289
+ "loss": 0.069,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 4.22,
294
+ "learning_rate": 1.1558213155117834e-05,
295
+ "loss": 0.0686,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 4.31,
300
+ "learning_rate": 1.1382342595849456e-05,
301
+ "loss": 0.0709,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 4.4,
306
+ "learning_rate": 1.1206472036581078e-05,
307
+ "loss": 0.0747,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 4.48,
312
+ "learning_rate": 1.1030601477312698e-05,
313
+ "loss": 0.0658,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 4.57,
318
+ "learning_rate": 1.085473091804432e-05,
319
+ "loss": 0.0732,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 4.66,
324
+ "learning_rate": 1.0678860358775942e-05,
325
+ "loss": 0.0715,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 4.75,
330
+ "learning_rate": 1.0502989799507564e-05,
331
+ "loss": 0.0778,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 4.84,
336
+ "learning_rate": 1.0327119240239184e-05,
337
+ "loss": 0.0724,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 4.92,
342
+ "learning_rate": 1.0151248680970806e-05,
343
+ "loss": 0.0722,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 5.01,
348
+ "learning_rate": 9.975378121702428e-06,
349
+ "loss": 0.0691,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 5.1,
354
+ "learning_rate": 9.79950756243405e-06,
355
+ "loss": 0.0485,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 5.19,
360
+ "learning_rate": 9.62363700316567e-06,
361
+ "loss": 0.0503,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 5.28,
366
+ "learning_rate": 9.447766443897292e-06,
367
+ "loss": 0.0536,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 5.36,
372
+ "learning_rate": 9.271895884628914e-06,
373
+ "loss": 0.0503,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 5.45,
378
+ "learning_rate": 9.096025325360536e-06,
379
+ "loss": 0.0452,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 5.54,
384
+ "learning_rate": 8.920154766092157e-06,
385
+ "loss": 0.0473,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 5.63,
390
+ "learning_rate": 8.744284206823778e-06,
391
+ "loss": 0.0488,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 5.72,
396
+ "learning_rate": 8.5684136475554e-06,
397
+ "loss": 0.0507,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 5.8,
402
+ "learning_rate": 8.392543088287022e-06,
403
+ "loss": 0.0467,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 5.89,
408
+ "learning_rate": 8.216672529018643e-06,
409
+ "loss": 0.0556,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 5.98,
414
+ "learning_rate": 8.040801969750265e-06,
415
+ "loss": 0.0481,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 6.07,
420
+ "learning_rate": 7.864931410481886e-06,
421
+ "loss": 0.0367,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 6.16,
426
+ "learning_rate": 7.689060851213508e-06,
427
+ "loss": 0.0329,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 6.24,
432
+ "learning_rate": 7.5131902919451295e-06,
433
+ "loss": 0.0294,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 6.33,
438
+ "learning_rate": 7.3373197326767506e-06,
439
+ "loss": 0.032,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 6.42,
444
+ "learning_rate": 7.1614491734083725e-06,
445
+ "loss": 0.0331,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 6.51,
450
+ "learning_rate": 6.985578614139994e-06,
451
+ "loss": 0.0274,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 6.6,
456
+ "learning_rate": 6.8097080548716155e-06,
457
+ "loss": 0.0385,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 6.68,
462
+ "learning_rate": 6.633837495603237e-06,
463
+ "loss": 0.0367,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 6.77,
468
+ "learning_rate": 6.4579669363348586e-06,
469
+ "loss": 0.0447,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 6.86,
474
+ "learning_rate": 6.28209637706648e-06,
475
+ "loss": 0.0433,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 6.95,
480
+ "learning_rate": 6.106225817798102e-06,
481
+ "loss": 0.034,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 7.03,
486
+ "learning_rate": 5.930355258529723e-06,
487
+ "loss": 0.0311,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 7.12,
492
+ "learning_rate": 5.754484699261345e-06,
493
+ "loss": 0.0189,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 7.21,
498
+ "learning_rate": 5.578614139992966e-06,
499
+ "loss": 0.02,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 7.3,
504
+ "learning_rate": 5.402743580724588e-06,
505
+ "loss": 0.0218,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 7.39,
510
+ "learning_rate": 5.226873021456209e-06,
511
+ "loss": 0.0217,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 7.47,
516
+ "learning_rate": 5.051002462187831e-06,
517
+ "loss": 0.0282,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 7.56,
522
+ "learning_rate": 4.875131902919452e-06,
523
+ "loss": 0.0267,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 7.65,
528
+ "learning_rate": 4.699261343651073e-06,
529
+ "loss": 0.026,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 7.74,
534
+ "learning_rate": 4.523390784382695e-06,
535
+ "loss": 0.0217,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 7.83,
540
+ "learning_rate": 4.347520225114316e-06,
541
+ "loss": 0.0194,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 7.91,
546
+ "learning_rate": 4.171649665845938e-06,
547
+ "loss": 0.026,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 8.0,
552
+ "learning_rate": 3.995779106577559e-06,
553
+ "loss": 0.0274,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 8.09,
558
+ "learning_rate": 3.819908547309181e-06,
559
+ "loss": 0.0107,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 8.18,
564
+ "learning_rate": 3.6440379880408023e-06,
565
+ "loss": 0.0103,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 8.27,
570
+ "learning_rate": 3.468167428772424e-06,
571
+ "loss": 0.0172,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 8.35,
576
+ "learning_rate": 3.2922968695040454e-06,
577
+ "loss": 0.0129,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 8.44,
582
+ "learning_rate": 3.116426310235667e-06,
583
+ "loss": 0.017,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 8.53,
588
+ "learning_rate": 2.9405557509672884e-06,
589
+ "loss": 0.0204,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 8.62,
594
+ "learning_rate": 2.76468519169891e-06,
595
+ "loss": 0.0128,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 8.71,
600
+ "learning_rate": 2.5888146324305314e-06,
601
+ "loss": 0.0213,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 8.79,
606
+ "learning_rate": 2.412944073162153e-06,
607
+ "loss": 0.0125,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 8.88,
612
+ "learning_rate": 2.2370735138937744e-06,
613
+ "loss": 0.0128,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 8.97,
618
+ "learning_rate": 2.061202954625396e-06,
619
+ "loss": 0.0182,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 9.06,
624
+ "learning_rate": 1.8853323953570175e-06,
625
+ "loss": 0.0116,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 9.15,
630
+ "learning_rate": 1.709461836088639e-06,
631
+ "loss": 0.0081,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 9.23,
636
+ "learning_rate": 1.5335912768202605e-06,
637
+ "loss": 0.0084,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 9.32,
642
+ "learning_rate": 1.357720717551882e-06,
643
+ "loss": 0.0076,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 9.41,
648
+ "learning_rate": 1.1818501582835035e-06,
649
+ "loss": 0.0089,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 9.5,
654
+ "learning_rate": 1.005979599015125e-06,
655
+ "loss": 0.0067,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 9.58,
660
+ "learning_rate": 8.301090397467465e-07,
661
+ "loss": 0.0066,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 9.67,
666
+ "learning_rate": 6.542384804783681e-07,
667
+ "loss": 0.0097,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 9.76,
672
+ "learning_rate": 4.783679212099895e-07,
673
+ "loss": 0.0099,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 9.85,
678
+ "learning_rate": 3.02497361941611e-07,
679
+ "loss": 0.0059,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 9.94,
684
+ "learning_rate": 1.2662680267323252e-07,
685
+ "loss": 0.0075,
686
+ "step": 56500
687
+ },
688
  {
689
  "epoch": 10.0,
690
+ "step": 56860,
691
+ "total_flos": 9.365103496606515e+17,
692
+ "train_runtime": 69350.928,
693
+ "train_samples_per_second": 52.464,
694
+ "train_steps_per_second": 0.82
 
695
  }
696
  ],
697
+ "max_steps": 56860,
698
  "num_train_epochs": 10,
699
+ "total_flos": 9.365103496606515e+17,
700
  "trial_name": null,
701
  "trial_params": null
702
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:80654956e83b29e2faf4498cab059199d52d2a9ad1fbf2339f1e6b3d3062857e
3
- size 2479
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9d87cd302a512898ad841afeff88ff5fd45499496cd61e69b97162c6837f041
3
+ size 2415