Model save
Browse files- all_results.json +18 -0
- eval_results.json +12 -0
- train_results.json +9 -0
- trainer_state.json +353 -0
all_results.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"eval_accuracy": 0.9768446160123027,
|
4 |
+
"eval_f1": 0.9388393674827523,
|
5 |
+
"eval_loss": 0.05124881863594055,
|
6 |
+
"eval_precision": 0.9288864055090761,
|
7 |
+
"eval_recall": 0.9490079303854105,
|
8 |
+
"eval_runtime": 122.2255,
|
9 |
+
"eval_samples": 20927,
|
10 |
+
"eval_samples_per_second": 171.216,
|
11 |
+
"eval_steps_per_second": 10.702,
|
12 |
+
"total_flos": 1.64925784065024e+16,
|
13 |
+
"train_loss": 0.0881471913733378,
|
14 |
+
"train_runtime": 3285.9718,
|
15 |
+
"train_samples": 167408,
|
16 |
+
"train_samples_per_second": 101.893,
|
17 |
+
"train_steps_per_second": 6.368
|
18 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"eval_accuracy": 0.9768446160123027,
|
4 |
+
"eval_f1": 0.9388393674827523,
|
5 |
+
"eval_loss": 0.05124881863594055,
|
6 |
+
"eval_precision": 0.9288864055090761,
|
7 |
+
"eval_recall": 0.9490079303854105,
|
8 |
+
"eval_runtime": 122.2255,
|
9 |
+
"eval_samples": 20927,
|
10 |
+
"eval_samples_per_second": 171.216,
|
11 |
+
"eval_steps_per_second": 10.702
|
12 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"total_flos": 1.64925784065024e+16,
|
4 |
+
"train_loss": 0.0881471913733378,
|
5 |
+
"train_runtime": 3285.9718,
|
6 |
+
"train_samples": 167408,
|
7 |
+
"train_samples_per_second": 101.893,
|
8 |
+
"train_steps_per_second": 6.368
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,353 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.05124881863594055,
|
3 |
+
"best_model_checkpoint": "my_Pytorch_pii_model/checkpoint-20926",
|
4 |
+
"epoch": 2.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 20926,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.04778744146038421,
|
13 |
+
"grad_norm": 1.7837579250335693,
|
14 |
+
"learning_rate": 1.997195242749148e-05,
|
15 |
+
"loss": 0.8081,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.09557488292076842,
|
20 |
+
"grad_norm": 2.2397186756134033,
|
21 |
+
"learning_rate": 1.9887742837017725e-05,
|
22 |
+
"loss": 0.2477,
|
23 |
+
"step": 1000
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.14336232438115262,
|
27 |
+
"grad_norm": 1.895089030265808,
|
28 |
+
"learning_rate": 1.9747845385097936e-05,
|
29 |
+
"loss": 0.1609,
|
30 |
+
"step": 1500
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.19114976584153684,
|
34 |
+
"grad_norm": 1.4316082000732422,
|
35 |
+
"learning_rate": 1.9553047975508295e-05,
|
36 |
+
"loss": 0.1269,
|
37 |
+
"step": 2000
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.23893720730192106,
|
41 |
+
"grad_norm": 0.5299736857414246,
|
42 |
+
"learning_rate": 1.9304447709107316e-05,
|
43 |
+
"loss": 0.1052,
|
44 |
+
"step": 2500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.28672464876230525,
|
48 |
+
"grad_norm": 1.0913931131362915,
|
49 |
+
"learning_rate": 1.9003444704953408e-05,
|
50 |
+
"loss": 0.0967,
|
51 |
+
"step": 3000
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.33451209022268946,
|
55 |
+
"grad_norm": 1.8749414682388306,
|
56 |
+
"learning_rate": 1.865248697176191e-05,
|
57 |
+
"loss": 0.0889,
|
58 |
+
"step": 3500
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.3822995316830737,
|
62 |
+
"grad_norm": 2.014411211013794,
|
63 |
+
"learning_rate": 1.825214512447262e-05,
|
64 |
+
"loss": 0.0843,
|
65 |
+
"step": 4000
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.4300869731434579,
|
69 |
+
"grad_norm": 0.6492653489112854,
|
70 |
+
"learning_rate": 1.780532711756845e-05,
|
71 |
+
"loss": 0.0812,
|
72 |
+
"step": 4500
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.4778744146038421,
|
76 |
+
"grad_norm": 1.5026785135269165,
|
77 |
+
"learning_rate": 1.7314549434302465e-05,
|
78 |
+
"loss": 0.0779,
|
79 |
+
"step": 5000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.5256618560642263,
|
83 |
+
"grad_norm": 0.5166222453117371,
|
84 |
+
"learning_rate": 1.6782576139241983e-05,
|
85 |
+
"loss": 0.0719,
|
86 |
+
"step": 5500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.5734492975246105,
|
90 |
+
"grad_norm": 0.9983482956886292,
|
91 |
+
"learning_rate": 1.62124033110307e-05,
|
92 |
+
"loss": 0.0746,
|
93 |
+
"step": 6000
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.6212367389849948,
|
97 |
+
"grad_norm": 0.8525393605232239,
|
98 |
+
"learning_rate": 1.5608485174626132e-05,
|
99 |
+
"loss": 0.0713,
|
100 |
+
"step": 6500
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.6690241804453789,
|
104 |
+
"grad_norm": 1.3193893432617188,
|
105 |
+
"learning_rate": 1.4971803628070286e-05,
|
106 |
+
"loss": 0.067,
|
107 |
+
"step": 7000
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.7168116219057632,
|
111 |
+
"grad_norm": 1.7481768131256104,
|
112 |
+
"learning_rate": 1.4307120836322233e-05,
|
113 |
+
"loss": 0.0677,
|
114 |
+
"step": 7500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.7645990633661474,
|
118 |
+
"grad_norm": 0.7293727993965149,
|
119 |
+
"learning_rate": 1.3618180299169794e-05,
|
120 |
+
"loss": 0.0649,
|
121 |
+
"step": 8000
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.8123865048265316,
|
125 |
+
"grad_norm": 0.4748859405517578,
|
126 |
+
"learning_rate": 1.2908862136252995e-05,
|
127 |
+
"loss": 0.0675,
|
128 |
+
"step": 8500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.8601739462869158,
|
132 |
+
"grad_norm": 0.9915094971656799,
|
133 |
+
"learning_rate": 1.218462628244987e-05,
|
134 |
+
"loss": 0.0648,
|
135 |
+
"step": 9000
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.9079613877473001,
|
139 |
+
"grad_norm": 2.1113200187683105,
|
140 |
+
"learning_rate": 1.1446650257897558e-05,
|
141 |
+
"loss": 0.0638,
|
142 |
+
"step": 9500
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.9557488292076842,
|
146 |
+
"grad_norm": 0.7165895700454712,
|
147 |
+
"learning_rate": 1.0700526685370893e-05,
|
148 |
+
"loss": 0.0615,
|
149 |
+
"step": 10000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.0,
|
153 |
+
"eval_accuracy": 0.9751704843504109,
|
154 |
+
"eval_f1": 0.9315204370483973,
|
155 |
+
"eval_loss": 0.05543896183371544,
|
156 |
+
"eval_precision": 0.9214454994188416,
|
157 |
+
"eval_recall": 0.9418181257135804,
|
158 |
+
"eval_runtime": 190.6513,
|
159 |
+
"eval_samples_per_second": 109.766,
|
160 |
+
"eval_steps_per_second": 6.861,
|
161 |
+
"step": 10463
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.0035362706680684,
|
165 |
+
"grad_norm": 0.7908827066421509,
|
166 |
+
"learning_rate": 9.950457739907535e-06,
|
167 |
+
"loss": 0.0591,
|
168 |
+
"step": 10500
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 1.0513237121284527,
|
172 |
+
"grad_norm": 0.8281264901161194,
|
173 |
+
"learning_rate": 9.200667816922936e-06,
|
174 |
+
"loss": 0.0559,
|
175 |
+
"step": 11000
|
176 |
+
},
|
177 |
+
{
|
178 |
+
"epoch": 1.099111153588837,
|
179 |
+
"grad_norm": 0.6375882625579834,
|
180 |
+
"learning_rate": 8.456863027173128e-06,
|
181 |
+
"loss": 0.0528,
|
182 |
+
"step": 11500
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.146898595049221,
|
186 |
+
"grad_norm": 0.8089446425437927,
|
187 |
+
"learning_rate": 7.720252707524596e-06,
|
188 |
+
"loss": 0.0561,
|
189 |
+
"step": 12000
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 1.1946860365096053,
|
193 |
+
"grad_norm": 0.5335150957107544,
|
194 |
+
"learning_rate": 6.9964819462513174e-06,
|
195 |
+
"loss": 0.0563,
|
196 |
+
"step": 12500
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 1.2424734779699895,
|
200 |
+
"grad_norm": 0.7552749514579773,
|
201 |
+
"learning_rate": 6.289627027145937e-06,
|
202 |
+
"loss": 0.053,
|
203 |
+
"step": 13000
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.2902609194303736,
|
207 |
+
"grad_norm": 1.121811866760254,
|
208 |
+
"learning_rate": 5.603668963817577e-06,
|
209 |
+
"loss": 0.0537,
|
210 |
+
"step": 13500
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 1.3380483608907578,
|
214 |
+
"grad_norm": 1.5103745460510254,
|
215 |
+
"learning_rate": 4.942471078585398e-06,
|
216 |
+
"loss": 0.0529,
|
217 |
+
"step": 14000
|
218 |
+
},
|
219 |
+
{
|
220 |
+
"epoch": 1.3858358023511421,
|
221 |
+
"grad_norm": 1.125802755355835,
|
222 |
+
"learning_rate": 4.309757244210676e-06,
|
223 |
+
"loss": 0.0558,
|
224 |
+
"step": 14500
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 1.4336232438115264,
|
228 |
+
"grad_norm": 2.0246198177337646,
|
229 |
+
"learning_rate": 3.709090911010067e-06,
|
230 |
+
"loss": 0.051,
|
231 |
+
"step": 15000
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 1.4814106852719107,
|
235 |
+
"grad_norm": 0.31454652547836304,
|
236 |
+
"learning_rate": 3.1449479986513345e-06,
|
237 |
+
"loss": 0.0503,
|
238 |
+
"step": 15500
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 1.5291981267322947,
|
242 |
+
"grad_norm": 1.1729323863983154,
|
243 |
+
"learning_rate": 2.618245735460083e-06,
|
244 |
+
"loss": 0.0496,
|
245 |
+
"step": 16000
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 1.576985568192679,
|
249 |
+
"grad_norm": 1.942895531654358,
|
250 |
+
"learning_rate": 2.1331175823046777e-06,
|
251 |
+
"loss": 0.0508,
|
252 |
+
"step": 16500
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 1.624773009653063,
|
256 |
+
"grad_norm": 0.7527912259101868,
|
257 |
+
"learning_rate": 1.692295785545267e-06,
|
258 |
+
"loss": 0.0474,
|
259 |
+
"step": 17000
|
260 |
+
},
|
261 |
+
{
|
262 |
+
"epoch": 1.6725604511134473,
|
263 |
+
"grad_norm": 0.9754999279975891,
|
264 |
+
"learning_rate": 1.299002907484831e-06,
|
265 |
+
"loss": 0.0505,
|
266 |
+
"step": 17500
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 1.7203478925738316,
|
270 |
+
"grad_norm": 1.4431662559509277,
|
271 |
+
"learning_rate": 9.538783949020436e-07,
|
272 |
+
"loss": 0.0507,
|
273 |
+
"step": 18000
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 1.7681353340342159,
|
277 |
+
"grad_norm": 0.9095632433891296,
|
278 |
+
"learning_rate": 6.602380824095455e-07,
|
279 |
+
"loss": 0.0484,
|
280 |
+
"step": 18500
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 1.8159227754946001,
|
284 |
+
"grad_norm": 0.47108370065689087,
|
285 |
+
"learning_rate": 4.185593937350141e-07,
|
286 |
+
"loss": 0.0503,
|
287 |
+
"step": 19000
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 1.8637102169549842,
|
291 |
+
"grad_norm": 1.9777565002441406,
|
292 |
+
"learning_rate": 2.3084346943755388e-07,
|
293 |
+
"loss": 0.0492,
|
294 |
+
"step": 19500
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.9114976584153685,
|
298 |
+
"grad_norm": 0.7806472182273865,
|
299 |
+
"learning_rate": 9.814752738334654e-08,
|
300 |
+
"loss": 0.0493,
|
301 |
+
"step": 20000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 1.9592850998757525,
|
305 |
+
"grad_norm": 0.42741918563842773,
|
306 |
+
"learning_rate": 2.1218912376697043e-08,
|
307 |
+
"loss": 0.051,
|
308 |
+
"step": 20500
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 2.0,
|
312 |
+
"eval_accuracy": 0.9768446160123027,
|
313 |
+
"eval_f1": 0.9388393674827523,
|
314 |
+
"eval_loss": 0.05124881863594055,
|
315 |
+
"eval_precision": 0.9288864055090761,
|
316 |
+
"eval_recall": 0.9490079303854105,
|
317 |
+
"eval_runtime": 115.4429,
|
318 |
+
"eval_samples_per_second": 181.276,
|
319 |
+
"eval_steps_per_second": 11.33,
|
320 |
+
"step": 20926
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 2.0,
|
324 |
+
"step": 20926,
|
325 |
+
"total_flos": 1.64925784065024e+16,
|
326 |
+
"train_loss": 0.0881471913733378,
|
327 |
+
"train_runtime": 3285.9718,
|
328 |
+
"train_samples_per_second": 101.893,
|
329 |
+
"train_steps_per_second": 6.368
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 2.0,
|
333 |
+
"eval_accuracy": 0.9768446160123027,
|
334 |
+
"eval_f1": 0.9388393674827523,
|
335 |
+
"eval_loss": 0.05124881863594055,
|
336 |
+
"eval_precision": 0.9288864055090761,
|
337 |
+
"eval_recall": 0.9490079303854105,
|
338 |
+
"eval_runtime": 122.2255,
|
339 |
+
"eval_samples_per_second": 171.216,
|
340 |
+
"eval_steps_per_second": 10.702,
|
341 |
+
"step": 20926
|
342 |
+
}
|
343 |
+
],
|
344 |
+
"logging_steps": 500,
|
345 |
+
"max_steps": 20926,
|
346 |
+
"num_input_tokens_seen": 0,
|
347 |
+
"num_train_epochs": 2,
|
348 |
+
"save_steps": 500,
|
349 |
+
"total_flos": 1.64925784065024e+16,
|
350 |
+
"train_batch_size": 16,
|
351 |
+
"trial_name": null,
|
352 |
+
"trial_params": null
|
353 |
+
}
|