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:e7e3fd16099426455efff4833878caa460e12cddbef8c5955642c3ab32d885b6
|
3 |
+
size 1711848436
|
checkpoint-12500/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8f50d8c0ef2e89b80e7aa4210d2b7c14af0e71a9d4cfa797b8fc1a99be05bc57
|
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 sha256:1ea02033f794b39b2796c8b06730a737bdd358bf661c368a409741348b1b636e
|
3 |
+
size 623
|
checkpoint-12500/trainer_state.json
ADDED
@@ -0,0 +1,408 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7920646667480469,
|
3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l17-l/checkpoint-12500",
|
4 |
+
"epoch": 1.3331911262798635,
|
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": 0.05332764505119454,
|
13 |
+
"grad_norm": 246.1711883544922,
|
14 |
+
"learning_rate": 9.822241183162683e-08,
|
15 |
+
"loss": 0.4515,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.05332764505119454,
|
20 |
+
"eval_loss": 1.3850886821746826,
|
21 |
+
"eval_runtime": 63.8439,
|
22 |
+
"eval_samples_per_second": 15.46,
|
23 |
+
"eval_steps_per_second": 1.942,
|
24 |
+
"step": 500
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 0.10665529010238908,
|
28 |
+
"grad_norm": 81.48503875732422,
|
29 |
+
"learning_rate": 9.64448236632537e-08,
|
30 |
+
"loss": 0.4148,
|
31 |
+
"step": 1000
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 0.10665529010238908,
|
35 |
+
"eval_loss": 1.284144401550293,
|
36 |
+
"eval_runtime": 63.853,
|
37 |
+
"eval_samples_per_second": 15.457,
|
38 |
+
"eval_steps_per_second": 1.942,
|
39 |
+
"step": 1000
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.1599829351535836,
|
43 |
+
"grad_norm": 718.367919921875,
|
44 |
+
"learning_rate": 9.466723549488054e-08,
|
45 |
+
"loss": 0.3281,
|
46 |
+
"step": 1500
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 0.1599829351535836,
|
50 |
+
"eval_loss": 1.2112621068954468,
|
51 |
+
"eval_runtime": 64.127,
|
52 |
+
"eval_samples_per_second": 15.391,
|
53 |
+
"eval_steps_per_second": 1.934,
|
54 |
+
"step": 1500
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 0.21331058020477817,
|
58 |
+
"grad_norm": 4.5923566818237305,
|
59 |
+
"learning_rate": 9.288964732650739e-08,
|
60 |
+
"loss": 0.2912,
|
61 |
+
"step": 2000
|
62 |
+
},
|
63 |
+
{
|
64 |
+
"epoch": 0.21331058020477817,
|
65 |
+
"eval_loss": 1.159011960029602,
|
66 |
+
"eval_runtime": 63.7062,
|
67 |
+
"eval_samples_per_second": 15.493,
|
68 |
+
"eval_steps_per_second": 1.946,
|
69 |
+
"step": 2000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.2666382252559727,
|
73 |
+
"grad_norm": 319.4483642578125,
|
74 |
+
"learning_rate": 9.111205915813424e-08,
|
75 |
+
"loss": 0.3073,
|
76 |
+
"step": 2500
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.2666382252559727,
|
80 |
+
"eval_loss": 1.1095446348190308,
|
81 |
+
"eval_runtime": 63.9925,
|
82 |
+
"eval_samples_per_second": 15.424,
|
83 |
+
"eval_steps_per_second": 1.938,
|
84 |
+
"step": 2500
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.3199658703071672,
|
88 |
+
"grad_norm": 0.00027541385497897863,
|
89 |
+
"learning_rate": 8.933447098976109e-08,
|
90 |
+
"loss": 0.2446,
|
91 |
+
"step": 3000
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 0.3199658703071672,
|
95 |
+
"eval_loss": 1.069318175315857,
|
96 |
+
"eval_runtime": 64.1107,
|
97 |
+
"eval_samples_per_second": 15.395,
|
98 |
+
"eval_steps_per_second": 1.934,
|
99 |
+
"step": 3000
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.37329351535836175,
|
103 |
+
"grad_norm": 5.2035698890686035,
|
104 |
+
"learning_rate": 8.755688282138793e-08,
|
105 |
+
"loss": 0.273,
|
106 |
+
"step": 3500
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.37329351535836175,
|
110 |
+
"eval_loss": 1.043074369430542,
|
111 |
+
"eval_runtime": 63.9447,
|
112 |
+
"eval_samples_per_second": 15.435,
|
113 |
+
"eval_steps_per_second": 1.939,
|
114 |
+
"step": 3500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.42662116040955633,
|
118 |
+
"grad_norm": 3.6792104244232178,
|
119 |
+
"learning_rate": 8.57792946530148e-08,
|
120 |
+
"loss": 0.193,
|
121 |
+
"step": 4000
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.42662116040955633,
|
125 |
+
"eval_loss": 1.0195527076721191,
|
126 |
+
"eval_runtime": 63.8076,
|
127 |
+
"eval_samples_per_second": 15.468,
|
128 |
+
"eval_steps_per_second": 1.943,
|
129 |
+
"step": 4000
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.47994880546075086,
|
133 |
+
"grad_norm": 111.2447280883789,
|
134 |
+
"learning_rate": 8.400170648464163e-08,
|
135 |
+
"loss": 0.2419,
|
136 |
+
"step": 4500
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.47994880546075086,
|
140 |
+
"eval_loss": 0.9946303963661194,
|
141 |
+
"eval_runtime": 63.8873,
|
142 |
+
"eval_samples_per_second": 15.449,
|
143 |
+
"eval_steps_per_second": 1.941,
|
144 |
+
"step": 4500
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.5332764505119454,
|
148 |
+
"grad_norm": 0.024256089702248573,
|
149 |
+
"learning_rate": 8.222411831626848e-08,
|
150 |
+
"loss": 0.2577,
|
151 |
+
"step": 5000
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 0.5332764505119454,
|
155 |
+
"eval_loss": 0.9679088592529297,
|
156 |
+
"eval_runtime": 62.896,
|
157 |
+
"eval_samples_per_second": 15.693,
|
158 |
+
"eval_steps_per_second": 1.972,
|
159 |
+
"step": 5000
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 0.58660409556314,
|
163 |
+
"grad_norm": 0.05403488501906395,
|
164 |
+
"learning_rate": 8.044653014789534e-08,
|
165 |
+
"loss": 0.2468,
|
166 |
+
"step": 5500
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 0.58660409556314,
|
170 |
+
"eval_loss": 0.9470070004463196,
|
171 |
+
"eval_runtime": 64.2367,
|
172 |
+
"eval_samples_per_second": 15.365,
|
173 |
+
"eval_steps_per_second": 1.93,
|
174 |
+
"step": 5500
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.6399317406143344,
|
178 |
+
"grad_norm": 56.23408508300781,
|
179 |
+
"learning_rate": 7.866894197952219e-08,
|
180 |
+
"loss": 0.2148,
|
181 |
+
"step": 6000
|
182 |
+
},
|
183 |
+
{
|
184 |
+
"epoch": 0.6399317406143344,
|
185 |
+
"eval_loss": 0.9275079369544983,
|
186 |
+
"eval_runtime": 62.7425,
|
187 |
+
"eval_samples_per_second": 15.731,
|
188 |
+
"eval_steps_per_second": 1.976,
|
189 |
+
"step": 6000
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.693259385665529,
|
193 |
+
"grad_norm": 335.7691955566406,
|
194 |
+
"learning_rate": 7.689135381114903e-08,
|
195 |
+
"loss": 0.179,
|
196 |
+
"step": 6500
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 0.693259385665529,
|
200 |
+
"eval_loss": 0.9107733964920044,
|
201 |
+
"eval_runtime": 63.711,
|
202 |
+
"eval_samples_per_second": 15.492,
|
203 |
+
"eval_steps_per_second": 1.946,
|
204 |
+
"step": 6500
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.7465870307167235,
|
208 |
+
"grad_norm": 0.0303946640342474,
|
209 |
+
"learning_rate": 7.511376564277589e-08,
|
210 |
+
"loss": 0.1992,
|
211 |
+
"step": 7000
|
212 |
+
},
|
213 |
+
{
|
214 |
+
"epoch": 0.7465870307167235,
|
215 |
+
"eval_loss": 0.8946455121040344,
|
216 |
+
"eval_runtime": 62.7465,
|
217 |
+
"eval_samples_per_second": 15.73,
|
218 |
+
"eval_steps_per_second": 1.976,
|
219 |
+
"step": 7000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.7999146757679181,
|
223 |
+
"grad_norm": 24.27122688293457,
|
224 |
+
"learning_rate": 7.333617747440273e-08,
|
225 |
+
"loss": 0.2264,
|
226 |
+
"step": 7500
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.7999146757679181,
|
230 |
+
"eval_loss": 0.8790715932846069,
|
231 |
+
"eval_runtime": 62.625,
|
232 |
+
"eval_samples_per_second": 15.76,
|
233 |
+
"eval_steps_per_second": 1.98,
|
234 |
+
"step": 7500
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 0.8532423208191127,
|
238 |
+
"grad_norm": 485.13018798828125,
|
239 |
+
"learning_rate": 7.155858930602957e-08,
|
240 |
+
"loss": 0.2349,
|
241 |
+
"step": 8000
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 0.8532423208191127,
|
245 |
+
"eval_loss": 0.8648595213890076,
|
246 |
+
"eval_runtime": 63.4718,
|
247 |
+
"eval_samples_per_second": 15.55,
|
248 |
+
"eval_steps_per_second": 1.954,
|
249 |
+
"step": 8000
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.9065699658703071,
|
253 |
+
"grad_norm": 39.18803787231445,
|
254 |
+
"learning_rate": 6.978100113765643e-08,
|
255 |
+
"loss": 0.1869,
|
256 |
+
"step": 8500
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 0.9065699658703071,
|
260 |
+
"eval_loss": 0.8521347641944885,
|
261 |
+
"eval_runtime": 62.649,
|
262 |
+
"eval_samples_per_second": 15.754,
|
263 |
+
"eval_steps_per_second": 1.979,
|
264 |
+
"step": 8500
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 0.9598976109215017,
|
268 |
+
"grad_norm": 15.91203784942627,
|
269 |
+
"learning_rate": 6.800341296928327e-08,
|
270 |
+
"loss": 0.1859,
|
271 |
+
"step": 9000
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 0.9598976109215017,
|
275 |
+
"eval_loss": 0.8395125269889832,
|
276 |
+
"eval_runtime": 63.1398,
|
277 |
+
"eval_samples_per_second": 15.632,
|
278 |
+
"eval_steps_per_second": 1.964,
|
279 |
+
"step": 9000
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.0132252559726962,
|
283 |
+
"grad_norm": 545.9375610351562,
|
284 |
+
"learning_rate": 6.622582480091012e-08,
|
285 |
+
"loss": 0.1943,
|
286 |
+
"step": 9500
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 1.0132252559726962,
|
290 |
+
"eval_loss": 0.8300319910049438,
|
291 |
+
"eval_runtime": 63.2047,
|
292 |
+
"eval_samples_per_second": 15.616,
|
293 |
+
"eval_steps_per_second": 1.962,
|
294 |
+
"step": 9500
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.0665529010238908,
|
298 |
+
"grad_norm": 451.04473876953125,
|
299 |
+
"learning_rate": 6.444823663253697e-08,
|
300 |
+
"loss": 0.1419,
|
301 |
+
"step": 10000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 1.0665529010238908,
|
305 |
+
"eval_loss": 0.8227387070655823,
|
306 |
+
"eval_runtime": 62.7357,
|
307 |
+
"eval_samples_per_second": 15.733,
|
308 |
+
"eval_steps_per_second": 1.977,
|
309 |
+
"step": 10000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 1.1198805460750854,
|
313 |
+
"grad_norm": 12.075934410095215,
|
314 |
+
"learning_rate": 6.267064846416382e-08,
|
315 |
+
"loss": 0.1558,
|
316 |
+
"step": 10500
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.1198805460750854,
|
320 |
+
"eval_loss": 0.8159177303314209,
|
321 |
+
"eval_runtime": 62.8297,
|
322 |
+
"eval_samples_per_second": 15.709,
|
323 |
+
"eval_steps_per_second": 1.974,
|
324 |
+
"step": 10500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 1.17320819112628,
|
328 |
+
"grad_norm": 363.5056457519531,
|
329 |
+
"learning_rate": 6.089306029579066e-08,
|
330 |
+
"loss": 0.15,
|
331 |
+
"step": 11000
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 1.17320819112628,
|
335 |
+
"eval_loss": 0.8096186518669128,
|
336 |
+
"eval_runtime": 63.3529,
|
337 |
+
"eval_samples_per_second": 15.579,
|
338 |
+
"eval_steps_per_second": 1.957,
|
339 |
+
"step": 11000
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.2265358361774745,
|
343 |
+
"grad_norm": 6.587007646885468e-06,
|
344 |
+
"learning_rate": 5.911547212741752e-08,
|
345 |
+
"loss": 0.1919,
|
346 |
+
"step": 11500
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.2265358361774745,
|
350 |
+
"eval_loss": 0.8023524284362793,
|
351 |
+
"eval_runtime": 62.9393,
|
352 |
+
"eval_samples_per_second": 15.682,
|
353 |
+
"eval_steps_per_second": 1.97,
|
354 |
+
"step": 11500
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 1.2798634812286689,
|
358 |
+
"grad_norm": 0.11630824208259583,
|
359 |
+
"learning_rate": 5.7337883959044365e-08,
|
360 |
+
"loss": 0.1298,
|
361 |
+
"step": 12000
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 1.2798634812286689,
|
365 |
+
"eval_loss": 0.7956322431564331,
|
366 |
+
"eval_runtime": 63.0726,
|
367 |
+
"eval_samples_per_second": 15.649,
|
368 |
+
"eval_steps_per_second": 1.966,
|
369 |
+
"step": 12000
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 1.3331911262798635,
|
373 |
+
"grad_norm": 0.0007865954539738595,
|
374 |
+
"learning_rate": 5.556029579067122e-08,
|
375 |
+
"loss": 0.1385,
|
376 |
+
"step": 12500
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 1.3331911262798635,
|
380 |
+
"eval_loss": 0.7920646667480469,
|
381 |
+
"eval_runtime": 63.3155,
|
382 |
+
"eval_samples_per_second": 15.589,
|
383 |
+
"eval_steps_per_second": 1.958,
|
384 |
+
"step": 12500
|
385 |
+
}
|
386 |
+
],
|
387 |
+
"logging_steps": 500,
|
388 |
+
"max_steps": 28128,
|
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": 4497698907452520.0,
|
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 sha256:52c8db40c27633302368150be011282e1650a78d29cdfc5e366df6966e8e6c70
|
3 |
+
size 4847
|