Training in progress, step 12500, checkpoint
Browse files
checkpoint-12500/config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/clip-vit-large-patch14-336",
|
3 |
+
"architectures": [
|
4 |
+
"CLIPModel"
|
5 |
+
],
|
6 |
+
"initializer_factor": 1.0,
|
7 |
+
"logit_scale_init_value": 2.6592,
|
8 |
+
"model_type": "clip",
|
9 |
+
"projection_dim": 768,
|
10 |
+
"text_config": {
|
11 |
+
"dropout": 0.0,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"intermediate_size": 3072,
|
14 |
+
"model_type": "clip_text_model",
|
15 |
+
"num_attention_heads": 12,
|
16 |
+
"projection_dim": 768
|
17 |
+
},
|
18 |
+
"torch_dtype": "float32",
|
19 |
+
"transformers_version": "4.45.0.dev0",
|
20 |
+
"vision_config": {
|
21 |
+
"dropout": 0.0,
|
22 |
+
"hidden_size": 1024,
|
23 |
+
"image_size": 336,
|
24 |
+
"intermediate_size": 4096,
|
25 |
+
"model_type": "clip_vision_model",
|
26 |
+
"num_attention_heads": 16,
|
27 |
+
"num_hidden_layers": 24,
|
28 |
+
"patch_size": 14,
|
29 |
+
"projection_dim": 768
|
30 |
+
}
|
31 |
+
}
|
checkpoint-12500/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:168c6e4d48ce2a1a14a7cb96c93b1788fe34028439a6bc614a235820a5aaa55d
|
3 |
+
size 1711848436
|
checkpoint-12500/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6540eb7421171e80e406056539cadd044a6d42ca3edf73487ff8ef54bdf7c22f
|
3 |
+
size 3424043887
|
checkpoint-12500/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:db030b30be6b93c99db6305a4a98097d55df00d054b7ea5674087844c194270b
|
3 |
+
size 14503
|
checkpoint-12500/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b0d0469b7fe82c6474dbeeccaa8d4f1e30aedc5009ab9364543400906510fa13
|
3 |
+
size 623
|
checkpoint-12500/trainer_state.json
ADDED
@@ -0,0 +1,408 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.6527238488197327,
|
3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-l/checkpoint-12500",
|
4 |
+
"epoch": 1.407182258246088,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 12500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.056287290329843524,
|
13 |
+
"grad_norm": 96.92656707763672,
|
14 |
+
"learning_rate": 2.9437127096701565e-07,
|
15 |
+
"loss": 0.3748,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.056287290329843524,
|
20 |
+
"eval_loss": 1.2226382493972778,
|
21 |
+
"eval_runtime": 131.2779,
|
22 |
+
"eval_samples_per_second": 15.037,
|
23 |
+
"eval_steps_per_second": 1.882,
|
24 |
+
"step": 500
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 0.11257458065968705,
|
28 |
+
"grad_norm": 402.31158447265625,
|
29 |
+
"learning_rate": 2.887425419340313e-07,
|
30 |
+
"loss": 0.3057,
|
31 |
+
"step": 1000
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 0.11257458065968705,
|
35 |
+
"eval_loss": 1.0702259540557861,
|
36 |
+
"eval_runtime": 131.0109,
|
37 |
+
"eval_samples_per_second": 15.067,
|
38 |
+
"eval_steps_per_second": 1.885,
|
39 |
+
"step": 1000
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.16886187098953057,
|
43 |
+
"grad_norm": 515.0911865234375,
|
44 |
+
"learning_rate": 2.831138129010469e-07,
|
45 |
+
"loss": 0.2239,
|
46 |
+
"step": 1500
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 0.16886187098953057,
|
50 |
+
"eval_loss": 0.9957238435745239,
|
51 |
+
"eval_runtime": 134.8377,
|
52 |
+
"eval_samples_per_second": 14.64,
|
53 |
+
"eval_steps_per_second": 1.832,
|
54 |
+
"step": 1500
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 0.2251491613193741,
|
58 |
+
"grad_norm": 3.1967105865478516,
|
59 |
+
"learning_rate": 2.774850838680626e-07,
|
60 |
+
"loss": 0.2229,
|
61 |
+
"step": 2000
|
62 |
+
},
|
63 |
+
{
|
64 |
+
"epoch": 0.2251491613193741,
|
65 |
+
"eval_loss": 0.9504629373550415,
|
66 |
+
"eval_runtime": 129.0169,
|
67 |
+
"eval_samples_per_second": 15.3,
|
68 |
+
"eval_steps_per_second": 1.914,
|
69 |
+
"step": 2000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.2814364516492176,
|
73 |
+
"grad_norm": 276.22998046875,
|
74 |
+
"learning_rate": 2.718563548350782e-07,
|
75 |
+
"loss": 0.2098,
|
76 |
+
"step": 2500
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.2814364516492176,
|
80 |
+
"eval_loss": 0.9006705284118652,
|
81 |
+
"eval_runtime": 129.1816,
|
82 |
+
"eval_samples_per_second": 15.281,
|
83 |
+
"eval_steps_per_second": 1.912,
|
84 |
+
"step": 2500
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.33772374197906113,
|
88 |
+
"grad_norm": 1105.317138671875,
|
89 |
+
"learning_rate": 2.6622762580209386e-07,
|
90 |
+
"loss": 0.1938,
|
91 |
+
"step": 3000
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 0.33772374197906113,
|
95 |
+
"eval_loss": 0.8782906532287598,
|
96 |
+
"eval_runtime": 128.6387,
|
97 |
+
"eval_samples_per_second": 15.345,
|
98 |
+
"eval_steps_per_second": 1.92,
|
99 |
+
"step": 3000
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.39401103230890466,
|
103 |
+
"grad_norm": 0.00043465400813147426,
|
104 |
+
"learning_rate": 2.605988967691095e-07,
|
105 |
+
"loss": 0.1688,
|
106 |
+
"step": 3500
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.39401103230890466,
|
110 |
+
"eval_loss": 0.8405746221542358,
|
111 |
+
"eval_runtime": 128.6989,
|
112 |
+
"eval_samples_per_second": 15.338,
|
113 |
+
"eval_steps_per_second": 1.919,
|
114 |
+
"step": 3500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.4502983226387482,
|
118 |
+
"grad_norm": 0.027426382526755333,
|
119 |
+
"learning_rate": 2.549701677361252e-07,
|
120 |
+
"loss": 0.1457,
|
121 |
+
"step": 4000
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.4502983226387482,
|
125 |
+
"eval_loss": 0.813768208026886,
|
126 |
+
"eval_runtime": 128.7209,
|
127 |
+
"eval_samples_per_second": 15.336,
|
128 |
+
"eval_steps_per_second": 1.919,
|
129 |
+
"step": 4000
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.5065856129685917,
|
133 |
+
"grad_norm": 0.0012935090344399214,
|
134 |
+
"learning_rate": 2.4934143870314085e-07,
|
135 |
+
"loss": 0.179,
|
136 |
+
"step": 4500
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.5065856129685917,
|
140 |
+
"eval_loss": 0.7965527176856995,
|
141 |
+
"eval_runtime": 128.3969,
|
142 |
+
"eval_samples_per_second": 15.374,
|
143 |
+
"eval_steps_per_second": 1.924,
|
144 |
+
"step": 4500
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.5628729032984352,
|
148 |
+
"grad_norm": 0.0017458726651966572,
|
149 |
+
"learning_rate": 2.4371270967015646e-07,
|
150 |
+
"loss": 0.1224,
|
151 |
+
"step": 5000
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 0.5628729032984352,
|
155 |
+
"eval_loss": 0.7788484692573547,
|
156 |
+
"eval_runtime": 128.4939,
|
157 |
+
"eval_samples_per_second": 15.363,
|
158 |
+
"eval_steps_per_second": 1.922,
|
159 |
+
"step": 5000
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 0.6191601936282788,
|
163 |
+
"grad_norm": 65.76844787597656,
|
164 |
+
"learning_rate": 2.380839806371721e-07,
|
165 |
+
"loss": 0.1551,
|
166 |
+
"step": 5500
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 0.6191601936282788,
|
170 |
+
"eval_loss": 0.7626588344573975,
|
171 |
+
"eval_runtime": 128.7302,
|
172 |
+
"eval_samples_per_second": 15.334,
|
173 |
+
"eval_steps_per_second": 1.919,
|
174 |
+
"step": 5500
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.6754474839581223,
|
178 |
+
"grad_norm": 5.014094829559326,
|
179 |
+
"learning_rate": 2.3245525160418776e-07,
|
180 |
+
"loss": 0.1721,
|
181 |
+
"step": 6000
|
182 |
+
},
|
183 |
+
{
|
184 |
+
"epoch": 0.6754474839581223,
|
185 |
+
"eval_loss": 0.7464810609817505,
|
186 |
+
"eval_runtime": 128.749,
|
187 |
+
"eval_samples_per_second": 15.332,
|
188 |
+
"eval_steps_per_second": 1.918,
|
189 |
+
"step": 6000
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.7317347742879657,
|
193 |
+
"grad_norm": 0.0022696161177009344,
|
194 |
+
"learning_rate": 2.268265225712034e-07,
|
195 |
+
"loss": 0.1532,
|
196 |
+
"step": 6500
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 0.7317347742879657,
|
200 |
+
"eval_loss": 0.7335842251777649,
|
201 |
+
"eval_runtime": 128.8165,
|
202 |
+
"eval_samples_per_second": 15.324,
|
203 |
+
"eval_steps_per_second": 1.917,
|
204 |
+
"step": 6500
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.7880220646178093,
|
208 |
+
"grad_norm": 5.508715730684344e-06,
|
209 |
+
"learning_rate": 2.2119779353821906e-07,
|
210 |
+
"loss": 0.1991,
|
211 |
+
"step": 7000
|
212 |
+
},
|
213 |
+
{
|
214 |
+
"epoch": 0.7880220646178093,
|
215 |
+
"eval_loss": 0.7244272232055664,
|
216 |
+
"eval_runtime": 128.9211,
|
217 |
+
"eval_samples_per_second": 15.312,
|
218 |
+
"eval_steps_per_second": 1.916,
|
219 |
+
"step": 7000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.8443093549476528,
|
223 |
+
"grad_norm": 0.30320531129837036,
|
224 |
+
"learning_rate": 2.155690645052347e-07,
|
225 |
+
"loss": 0.1551,
|
226 |
+
"step": 7500
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.8443093549476528,
|
230 |
+
"eval_loss": 0.718368411064148,
|
231 |
+
"eval_runtime": 128.8149,
|
232 |
+
"eval_samples_per_second": 15.324,
|
233 |
+
"eval_steps_per_second": 1.917,
|
234 |
+
"step": 7500
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 0.9005966452774964,
|
238 |
+
"grad_norm": 1.9788849385804497e-05,
|
239 |
+
"learning_rate": 2.0994033547225037e-07,
|
240 |
+
"loss": 0.1439,
|
241 |
+
"step": 8000
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 0.9005966452774964,
|
245 |
+
"eval_loss": 0.7040167450904846,
|
246 |
+
"eval_runtime": 128.4921,
|
247 |
+
"eval_samples_per_second": 15.363,
|
248 |
+
"eval_steps_per_second": 1.922,
|
249 |
+
"step": 8000
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.9568839356073399,
|
253 |
+
"grad_norm": 1.0750063665909693e-06,
|
254 |
+
"learning_rate": 2.04311606439266e-07,
|
255 |
+
"loss": 0.1361,
|
256 |
+
"step": 8500
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 0.9568839356073399,
|
260 |
+
"eval_loss": 0.6983720660209656,
|
261 |
+
"eval_runtime": 128.5929,
|
262 |
+
"eval_samples_per_second": 15.351,
|
263 |
+
"eval_steps_per_second": 1.921,
|
264 |
+
"step": 8500
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 1.0131712259371835,
|
268 |
+
"grad_norm": 187.1415557861328,
|
269 |
+
"learning_rate": 1.9868287740628167e-07,
|
270 |
+
"loss": 0.1144,
|
271 |
+
"step": 9000
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 1.0131712259371835,
|
275 |
+
"eval_loss": 0.6901102066040039,
|
276 |
+
"eval_runtime": 128.5322,
|
277 |
+
"eval_samples_per_second": 15.358,
|
278 |
+
"eval_steps_per_second": 1.922,
|
279 |
+
"step": 9000
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.069458516267027,
|
283 |
+
"grad_norm": 0.007205183617770672,
|
284 |
+
"learning_rate": 1.9305414837329728e-07,
|
285 |
+
"loss": 0.0643,
|
286 |
+
"step": 9500
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 1.069458516267027,
|
290 |
+
"eval_loss": 0.6831753253936768,
|
291 |
+
"eval_runtime": 128.59,
|
292 |
+
"eval_samples_per_second": 15.351,
|
293 |
+
"eval_steps_per_second": 1.921,
|
294 |
+
"step": 9500
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.1257458065968704,
|
298 |
+
"grad_norm": 0.4961595833301544,
|
299 |
+
"learning_rate": 1.8742541934031294e-07,
|
300 |
+
"loss": 0.0773,
|
301 |
+
"step": 10000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 1.1257458065968704,
|
305 |
+
"eval_loss": 0.6732445955276489,
|
306 |
+
"eval_runtime": 128.732,
|
307 |
+
"eval_samples_per_second": 15.334,
|
308 |
+
"eval_steps_per_second": 1.919,
|
309 |
+
"step": 10000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 1.1820330969267139,
|
313 |
+
"grad_norm": 0.0009373470675200224,
|
314 |
+
"learning_rate": 1.817966903073286e-07,
|
315 |
+
"loss": 0.0814,
|
316 |
+
"step": 10500
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.1820330969267139,
|
320 |
+
"eval_loss": 0.6723716259002686,
|
321 |
+
"eval_runtime": 126.1133,
|
322 |
+
"eval_samples_per_second": 15.653,
|
323 |
+
"eval_steps_per_second": 1.959,
|
324 |
+
"step": 10500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 1.2383203872565574,
|
328 |
+
"grad_norm": 2.7500348096509697e-06,
|
329 |
+
"learning_rate": 1.7616796127434424e-07,
|
330 |
+
"loss": 0.0924,
|
331 |
+
"step": 11000
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 1.2383203872565574,
|
335 |
+
"eval_loss": 0.6679245233535767,
|
336 |
+
"eval_runtime": 126.0452,
|
337 |
+
"eval_samples_per_second": 15.661,
|
338 |
+
"eval_steps_per_second": 1.96,
|
339 |
+
"step": 11000
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.294607677586401,
|
343 |
+
"grad_norm": 12.024052619934082,
|
344 |
+
"learning_rate": 1.705392322413599e-07,
|
345 |
+
"loss": 0.091,
|
346 |
+
"step": 11500
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.294607677586401,
|
350 |
+
"eval_loss": 0.6659140586853027,
|
351 |
+
"eval_runtime": 129.0611,
|
352 |
+
"eval_samples_per_second": 15.295,
|
353 |
+
"eval_steps_per_second": 1.914,
|
354 |
+
"step": 11500
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 1.3508949679162445,
|
358 |
+
"grad_norm": 0.0008855258929543197,
|
359 |
+
"learning_rate": 1.6491050320837554e-07,
|
360 |
+
"loss": 0.0978,
|
361 |
+
"step": 12000
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 1.3508949679162445,
|
365 |
+
"eval_loss": 0.6577230095863342,
|
366 |
+
"eval_runtime": 126.2858,
|
367 |
+
"eval_samples_per_second": 15.631,
|
368 |
+
"eval_steps_per_second": 1.956,
|
369 |
+
"step": 12000
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 1.407182258246088,
|
373 |
+
"grad_norm": 0.0003772446943912655,
|
374 |
+
"learning_rate": 1.592817741753912e-07,
|
375 |
+
"loss": 0.0598,
|
376 |
+
"step": 12500
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 1.407182258246088,
|
380 |
+
"eval_loss": 0.6527238488197327,
|
381 |
+
"eval_runtime": 125.8972,
|
382 |
+
"eval_samples_per_second": 15.679,
|
383 |
+
"eval_steps_per_second": 1.962,
|
384 |
+
"step": 12500
|
385 |
+
}
|
386 |
+
],
|
387 |
+
"logging_steps": 500,
|
388 |
+
"max_steps": 26649,
|
389 |
+
"num_input_tokens_seen": 0,
|
390 |
+
"num_train_epochs": 3,
|
391 |
+
"save_steps": 500,
|
392 |
+
"stateful_callbacks": {
|
393 |
+
"TrainerControl": {
|
394 |
+
"args": {
|
395 |
+
"should_epoch_stop": false,
|
396 |
+
"should_evaluate": false,
|
397 |
+
"should_log": false,
|
398 |
+
"should_save": true,
|
399 |
+
"should_training_stop": false
|
400 |
+
},
|
401 |
+
"attributes": {}
|
402 |
+
}
|
403 |
+
},
|
404 |
+
"total_flos": 4497519028276890.0,
|
405 |
+
"train_batch_size": 2,
|
406 |
+
"trial_name": null,
|
407 |
+
"trial_params": null
|
408 |
+
}
|
checkpoint-12500/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:074eb2b3aae1f3310fb0b27d83d259baa7ad74a32c8433c81af819e28558f6b0
|
3 |
+
size 4847
|