Model save
Browse files- README.md +4 -9
- all_results.json +5 -10
- model-00001-of-00004.safetensors +1 -1
- model-00002-of-00004.safetensors +1 -1
- model-00003-of-00004.safetensors +1 -1
- model-00004-of-00004.safetensors +1 -1
- runs/Jun14_19-49-54_action-graph-trainer/events.out.tfevents.1718394615.action-graph-trainer.710510.0 +2 -2
- train_results.json +5 -5
- trainer_state.json +258 -258
README.md
CHANGED
@@ -2,16 +2,11 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: Qwen/Qwen2-7B
|
4 |
tags:
|
5 |
-
- alignment-handbook
|
6 |
- trl
|
7 |
- sft
|
8 |
- generated_from_trainer
|
9 |
-
- trl
|
10 |
-
- sft
|
11 |
-
- alignment-handbook
|
12 |
-
- generated_from_trainer
|
13 |
datasets:
|
14 |
-
-
|
15 |
model-index:
|
16 |
- name: zephyr-qwen2-7b-sft
|
17 |
results: []
|
@@ -22,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
22 |
|
23 |
# zephyr-qwen2-7b-sft
|
24 |
|
25 |
-
This model is a fine-tuned version of [Qwen/Qwen2-7B](https://huggingface.co/Qwen/Qwen2-7B) on the
|
26 |
It achieves the following results on the evaluation set:
|
27 |
-
- Loss: 1.
|
28 |
|
29 |
## Model description
|
30 |
|
@@ -61,7 +56,7 @@ The following hyperparameters were used during training:
|
|
61 |
|
62 |
| Training Loss | Epoch | Step | Validation Loss |
|
63 |
|:-------------:|:-----:|:----:|:---------------:|
|
64 |
-
| 1.0627 | 1.0 | 956 | 1.
|
65 |
|
66 |
|
67 |
### Framework versions
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: Qwen/Qwen2-7B
|
4 |
tags:
|
|
|
5 |
- trl
|
6 |
- sft
|
7 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
8 |
datasets:
|
9 |
+
- generator
|
10 |
model-index:
|
11 |
- name: zephyr-qwen2-7b-sft
|
12 |
results: []
|
|
|
17 |
|
18 |
# zephyr-qwen2-7b-sft
|
19 |
|
20 |
+
This model is a fine-tuned version of [Qwen/Qwen2-7B](https://huggingface.co/Qwen/Qwen2-7B) on the generator dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 1.0646
|
23 |
|
24 |
## Model description
|
25 |
|
|
|
56 |
|
57 |
| Training Loss | Epoch | Step | Validation Loss |
|
58 |
|:-------------:|:-----:|:----:|:---------------:|
|
59 |
+
| 1.0627 | 1.0 | 956 | 1.0646 |
|
60 |
|
61 |
|
62 |
### Framework versions
|
all_results.json
CHANGED
@@ -1,14 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"
|
6 |
-
"eval_samples_per_second": 47.523,
|
7 |
-
"eval_steps_per_second": 0.744,
|
8 |
-
"total_flos": 500662995517440.0,
|
9 |
-
"train_loss": 0.06220405869902926,
|
10 |
-
"train_runtime": 877.8841,
|
11 |
"train_samples": 207864,
|
12 |
-
"train_samples_per_second":
|
13 |
-
"train_steps_per_second":
|
14 |
}
|
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
+
"total_flos": 500925122740224.0,
|
4 |
+
"train_loss": 1.0766765792000743,
|
5 |
+
"train_runtime": 10814.9645,
|
|
|
|
|
|
|
|
|
|
|
6 |
"train_samples": 207864,
|
7 |
+
"train_samples_per_second": 11.312,
|
8 |
+
"train_steps_per_second": 0.088
|
9 |
}
|
model-00001-of-00004.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4877660776
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b894e073d5831d48eadeb6fb65c8d6567b6f104857b13a3749f61753a309b8a8
|
3 |
size 4877660776
|
model-00002-of-00004.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4932751008
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e49794c679857c0943180fa9b1762d1e85d2db2bc0a08b5b89971b77a8fd9ae9
|
3 |
size 4932751008
|
model-00003-of-00004.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4330865200
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b99ccc1e751ad90605a4da706378602eb98496a62dc5567a3f13ab6c243c4919
|
3 |
size 4330865200
|
model-00004-of-00004.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1089994880
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7904dd9e655204dbefa882b077d61a3c8447d69646fd52fd93aec990d3fad7e7
|
3 |
size 1089994880
|
runs/Jun14_19-49-54_action-graph-trainer/events.out.tfevents.1718394615.action-graph-trainer.710510.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3299937ae81515a2ced16997b8e6ccf79122f09c316c6d8221ad16da8f1fee7c
|
3 |
+
size 45894
|
train_results.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
-
"total_flos":
|
4 |
-
"train_loss":
|
5 |
-
"train_runtime":
|
6 |
"train_samples": 207864,
|
7 |
-
"train_samples_per_second":
|
8 |
-
"train_steps_per_second":
|
9 |
}
|
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
+
"total_flos": 500925122740224.0,
|
4 |
+
"train_loss": 1.0766765792000743,
|
5 |
+
"train_runtime": 10814.9645,
|
6 |
"train_samples": 207864,
|
7 |
+
"train_samples_per_second": 11.312,
|
8 |
+
"train_steps_per_second": 0.088
|
9 |
}
|
trainer_state.json
CHANGED
@@ -10,1364 +10,1364 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.0010460251046025104,
|
13 |
-
"grad_norm": 4.
|
14 |
"learning_rate": 2.0833333333333333e-07,
|
15 |
"loss": 1.3092,
|
16 |
"step": 1
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.005230125523012552,
|
20 |
-
"grad_norm": 4.
|
21 |
"learning_rate": 1.0416666666666667e-06,
|
22 |
"loss": 1.2901,
|
23 |
"step": 5
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.010460251046025104,
|
27 |
-
"grad_norm": 4.
|
28 |
"learning_rate": 2.0833333333333334e-06,
|
29 |
"loss": 1.2488,
|
30 |
"step": 10
|
31 |
},
|
32 |
{
|
33 |
"epoch": 0.015690376569037656,
|
34 |
-
"grad_norm": 1.
|
35 |
"learning_rate": 3.125e-06,
|
36 |
-
"loss": 1.
|
37 |
"step": 15
|
38 |
},
|
39 |
{
|
40 |
"epoch": 0.02092050209205021,
|
41 |
-
"grad_norm": 0.
|
42 |
"learning_rate": 4.166666666666667e-06,
|
43 |
"loss": 1.1238,
|
44 |
"step": 20
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.02615062761506276,
|
48 |
-
"grad_norm": 0.
|
49 |
"learning_rate": 5.208333333333334e-06,
|
50 |
"loss": 1.0882,
|
51 |
"step": 25
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.03138075313807531,
|
55 |
-
"grad_norm": 0.
|
56 |
"learning_rate": 6.25e-06,
|
57 |
"loss": 1.0972,
|
58 |
"step": 30
|
59 |
},
|
60 |
{
|
61 |
"epoch": 0.036610878661087864,
|
62 |
-
"grad_norm": 0.
|
63 |
"learning_rate": 7.291666666666667e-06,
|
64 |
"loss": 1.0967,
|
65 |
"step": 35
|
66 |
},
|
67 |
{
|
68 |
"epoch": 0.04184100418410042,
|
69 |
-
"grad_norm": 0.
|
70 |
"learning_rate": 8.333333333333334e-06,
|
71 |
"loss": 1.0806,
|
72 |
"step": 40
|
73 |
},
|
74 |
{
|
75 |
"epoch": 0.04707112970711297,
|
76 |
-
"grad_norm": 0.
|
77 |
"learning_rate": 9.375000000000001e-06,
|
78 |
"loss": 1.0799,
|
79 |
"step": 45
|
80 |
},
|
81 |
{
|
82 |
"epoch": 0.05230125523012552,
|
83 |
-
"grad_norm": 0.
|
84 |
"learning_rate": 1.0416666666666668e-05,
|
85 |
"loss": 1.0776,
|
86 |
"step": 50
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.05753138075313807,
|
90 |
-
"grad_norm": 0.
|
91 |
"learning_rate": 1.1458333333333333e-05,
|
92 |
"loss": 1.0899,
|
93 |
"step": 55
|
94 |
},
|
95 |
{
|
96 |
"epoch": 0.06276150627615062,
|
97 |
-
"grad_norm": 0.
|
98 |
"learning_rate": 1.25e-05,
|
99 |
-
"loss": 1.
|
100 |
"step": 60
|
101 |
},
|
102 |
{
|
103 |
"epoch": 0.06799163179916318,
|
104 |
-
"grad_norm": 0.
|
105 |
"learning_rate": 1.3541666666666668e-05,
|
106 |
"loss": 1.0822,
|
107 |
"step": 65
|
108 |
},
|
109 |
{
|
110 |
"epoch": 0.07322175732217573,
|
111 |
-
"grad_norm": 0.
|
112 |
"learning_rate": 1.4583333333333333e-05,
|
113 |
-
"loss": 1.
|
114 |
"step": 70
|
115 |
},
|
116 |
{
|
117 |
"epoch": 0.07845188284518828,
|
118 |
-
"grad_norm": 0.
|
119 |
"learning_rate": 1.5625e-05,
|
120 |
"loss": 1.094,
|
121 |
"step": 75
|
122 |
},
|
123 |
{
|
124 |
"epoch": 0.08368200836820083,
|
125 |
-
"grad_norm": 0.
|
126 |
"learning_rate": 1.6666666666666667e-05,
|
127 |
"loss": 1.0929,
|
128 |
"step": 80
|
129 |
},
|
130 |
{
|
131 |
"epoch": 0.08891213389121339,
|
132 |
-
"grad_norm": 0.
|
133 |
"learning_rate": 1.7708333333333335e-05,
|
134 |
"loss": 1.1021,
|
135 |
"step": 85
|
136 |
},
|
137 |
{
|
138 |
"epoch": 0.09414225941422594,
|
139 |
-
"grad_norm": 0.
|
140 |
"learning_rate": 1.8750000000000002e-05,
|
141 |
"loss": 1.0856,
|
142 |
"step": 90
|
143 |
},
|
144 |
{
|
145 |
"epoch": 0.09937238493723849,
|
146 |
-
"grad_norm": 0.
|
147 |
"learning_rate": 1.979166666666667e-05,
|
148 |
"loss": 1.0862,
|
149 |
"step": 95
|
150 |
},
|
151 |
{
|
152 |
"epoch": 0.10460251046025104,
|
153 |
-
"grad_norm": 0.
|
154 |
"learning_rate": 1.9998932457674904e-05,
|
155 |
"loss": 1.0933,
|
156 |
"step": 100
|
157 |
},
|
158 |
{
|
159 |
"epoch": 0.1098326359832636,
|
160 |
-
"grad_norm": 0.
|
161 |
"learning_rate": 1.999459595761682e-05,
|
162 |
"loss": 1.1147,
|
163 |
"step": 105
|
164 |
},
|
165 |
{
|
166 |
"epoch": 0.11506276150627615,
|
167 |
-
"grad_norm": 0.
|
168 |
"learning_rate": 1.9986925223989665e-05,
|
169 |
"loss": 1.093,
|
170 |
"step": 110
|
171 |
},
|
172 |
{
|
173 |
"epoch": 0.1202928870292887,
|
174 |
-
"grad_norm": 0.
|
175 |
"learning_rate": 1.9975922815778082e-05,
|
176 |
"loss": 1.0831,
|
177 |
"step": 115
|
178 |
},
|
179 |
{
|
180 |
"epoch": 0.12552301255230125,
|
181 |
-
"grad_norm": 0.
|
182 |
"learning_rate": 1.996159240342547e-05,
|
183 |
"loss": 1.0825,
|
184 |
"step": 120
|
185 |
},
|
186 |
{
|
187 |
"epoch": 0.1307531380753138,
|
188 |
-
"grad_norm": 0.
|
189 |
"learning_rate": 1.9943938767609515e-05,
|
190 |
-
"loss": 1.
|
191 |
"step": 125
|
192 |
},
|
193 |
{
|
194 |
"epoch": 0.13598326359832635,
|
195 |
-
"grad_norm": 0.
|
196 |
"learning_rate": 1.9922967797647357e-05,
|
197 |
"loss": 1.0905,
|
198 |
"step": 130
|
199 |
},
|
200 |
{
|
201 |
"epoch": 0.1412133891213389,
|
202 |
-
"grad_norm": 0.
|
203 |
"learning_rate": 1.9898686489530852e-05,
|
204 |
"loss": 1.0974,
|
205 |
"step": 135
|
206 |
},
|
207 |
{
|
208 |
"epoch": 0.14644351464435146,
|
209 |
-
"grad_norm": 0.
|
210 |
"learning_rate": 1.9871102943592717e-05,
|
211 |
"loss": 1.0731,
|
212 |
"step": 140
|
213 |
},
|
214 |
{
|
215 |
"epoch": 0.151673640167364,
|
216 |
-
"grad_norm": 0.
|
217 |
"learning_rate": 1.9840226361804223e-05,
|
218 |
"loss": 1.1055,
|
219 |
"step": 145
|
220 |
},
|
221 |
{
|
222 |
"epoch": 0.15690376569037656,
|
223 |
-
"grad_norm": 0.
|
224 |
"learning_rate": 1.9806067044705375e-05,
|
225 |
-
"loss": 1.
|
226 |
"step": 150
|
227 |
},
|
228 |
{
|
229 |
"epoch": 0.16213389121338911,
|
230 |
-
"grad_norm": 0.
|
231 |
"learning_rate": 1.9768636387968616e-05,
|
232 |
"loss": 1.1064,
|
233 |
"step": 155
|
234 |
},
|
235 |
{
|
236 |
"epoch": 0.16736401673640167,
|
237 |
-
"grad_norm": 0.
|
238 |
"learning_rate": 1.9727946878597193e-05,
|
239 |
"loss": 1.0791,
|
240 |
"step": 160
|
241 |
},
|
242 |
{
|
243 |
"epoch": 0.17259414225941422,
|
244 |
-
"grad_norm": 0.
|
245 |
"learning_rate": 1.968401209075946e-05,
|
246 |
"loss": 1.0839,
|
247 |
"step": 165
|
248 |
},
|
249 |
{
|
250 |
"epoch": 0.17782426778242677,
|
251 |
-
"grad_norm": 0.
|
252 |
"learning_rate": 1.963684668126046e-05,
|
253 |
-
"loss": 1.
|
254 |
"step": 170
|
255 |
},
|
256 |
{
|
257 |
"epoch": 0.18305439330543932,
|
258 |
-
"grad_norm": 0.
|
259 |
"learning_rate": 1.9586466384652366e-05,
|
260 |
"loss": 1.0847,
|
261 |
"step": 175
|
262 |
},
|
263 |
{
|
264 |
"epoch": 0.18828451882845187,
|
265 |
-
"grad_norm": 0.
|
266 |
"learning_rate": 1.9532888007985408e-05,
|
267 |
-
"loss": 1.
|
268 |
"step": 180
|
269 |
},
|
270 |
{
|
271 |
"epoch": 0.19351464435146443,
|
272 |
-
"grad_norm": 0.
|
273 |
"learning_rate": 1.9476129425200935e-05,
|
274 |
-
"loss": 1.
|
275 |
"step": 185
|
276 |
},
|
277 |
{
|
278 |
"epoch": 0.19874476987447698,
|
279 |
-
"grad_norm": 0.
|
280 |
"learning_rate": 1.9416209571168648e-05,
|
281 |
"loss": 1.089,
|
282 |
"step": 190
|
283 |
},
|
284 |
{
|
285 |
"epoch": 0.20397489539748953,
|
286 |
-
"grad_norm": 0.
|
287 |
"learning_rate": 1.9353148435369816e-05,
|
288 |
"loss": 1.0906,
|
289 |
"step": 195
|
290 |
},
|
291 |
{
|
292 |
"epoch": 0.20920502092050208,
|
293 |
-
"grad_norm": 0.
|
294 |
"learning_rate": 1.9286967055228744e-05,
|
295 |
"loss": 1.093,
|
296 |
"step": 200
|
297 |
},
|
298 |
{
|
299 |
"epoch": 0.21443514644351463,
|
300 |
-
"grad_norm": 0.
|
301 |
"learning_rate": 1.9217687509094596e-05,
|
302 |
"loss": 1.0759,
|
303 |
"step": 205
|
304 |
},
|
305 |
{
|
306 |
"epoch": 0.2196652719665272,
|
307 |
-
"grad_norm": 0.
|
308 |
"learning_rate": 1.9145332908875984e-05,
|
309 |
"loss": 1.0911,
|
310 |
"step": 210
|
311 |
},
|
312 |
{
|
313 |
"epoch": 0.22489539748953974,
|
314 |
-
"grad_norm": 0.
|
315 |
"learning_rate": 1.9069927392330742e-05,
|
316 |
"loss": 1.1017,
|
317 |
"step": 215
|
318 |
},
|
319 |
{
|
320 |
"epoch": 0.2301255230125523,
|
321 |
-
"grad_norm": 0.
|
322 |
"learning_rate": 1.89914961150135e-05,
|
323 |
"loss": 1.082,
|
324 |
"step": 220
|
325 |
},
|
326 |
{
|
327 |
"epoch": 0.23535564853556484,
|
328 |
-
"grad_norm": 0.
|
329 |
"learning_rate": 1.891006524188368e-05,
|
330 |
"loss": 1.0797,
|
331 |
"step": 225
|
332 |
},
|
333 |
{
|
334 |
"epoch": 0.2405857740585774,
|
335 |
-
"grad_norm": 0.
|
336 |
"learning_rate": 1.8825661938576784e-05,
|
337 |
"loss": 1.085,
|
338 |
"step": 230
|
339 |
},
|
340 |
{
|
341 |
"epoch": 0.24581589958158995,
|
342 |
-
"grad_norm": 0.
|
343 |
"learning_rate": 1.8738314362341828e-05,
|
344 |
-
"loss": 1.
|
345 |
"step": 235
|
346 |
},
|
347 |
{
|
348 |
"epoch": 0.2510460251046025,
|
349 |
-
"grad_norm": 0.
|
350 |
"learning_rate": 1.864805165264799e-05,
|
351 |
-
"loss": 1.
|
352 |
"step": 240
|
353 |
},
|
354 |
{
|
355 |
"epoch": 0.25627615062761505,
|
356 |
-
"grad_norm": 0.
|
357 |
"learning_rate": 1.8554903921463578e-05,
|
358 |
"loss": 1.0889,
|
359 |
"step": 245
|
360 |
},
|
361 |
{
|
362 |
"epoch": 0.2615062761506276,
|
363 |
-
"grad_norm": 0.
|
364 |
"learning_rate": 1.8458902243210558e-05,
|
365 |
"loss": 1.0822,
|
366 |
"step": 250
|
367 |
},
|
368 |
{
|
369 |
"epoch": 0.26673640167364016,
|
370 |
-
"grad_norm": 0.
|
371 |
"learning_rate": 1.8360078644398028e-05,
|
372 |
"loss": 1.0788,
|
373 |
"step": 255
|
374 |
},
|
375 |
{
|
376 |
"epoch": 0.2719665271966527,
|
377 |
-
"grad_norm": 0.
|
378 |
"learning_rate": 1.8258466092938042e-05,
|
379 |
"loss": 1.101,
|
380 |
"step": 260
|
381 |
},
|
382 |
{
|
383 |
"epoch": 0.27719665271966526,
|
384 |
-
"grad_norm": 0.
|
385 |
"learning_rate": 1.8154098487147405e-05,
|
386 |
"loss": 1.0776,
|
387 |
"step": 265
|
388 |
},
|
389 |
{
|
390 |
"epoch": 0.2824267782426778,
|
391 |
-
"grad_norm": 0.
|
392 |
"learning_rate": 1.8047010644439074e-05,
|
393 |
"loss": 1.0838,
|
394 |
"step": 270
|
395 |
},
|
396 |
{
|
397 |
"epoch": 0.28765690376569036,
|
398 |
-
"grad_norm": 0.
|
399 |
"learning_rate": 1.7937238289706927e-05,
|
400 |
-
"loss": 1.
|
401 |
"step": 275
|
402 |
},
|
403 |
{
|
404 |
"epoch": 0.2928870292887029,
|
405 |
-
"grad_norm": 0.
|
406 |
"learning_rate": 1.7824818043407828e-05,
|
407 |
-
"loss": 1.
|
408 |
"step": 280
|
409 |
},
|
410 |
{
|
411 |
"epoch": 0.29811715481171547,
|
412 |
-
"grad_norm": 0.
|
413 |
"learning_rate": 1.7709787409344893e-05,
|
414 |
-
"loss": 1.
|
415 |
"step": 285
|
416 |
},
|
417 |
{
|
418 |
"epoch": 0.303347280334728,
|
419 |
-
"grad_norm": 0.
|
420 |
"learning_rate": 1.75921847621561e-05,
|
421 |
"loss": 1.0863,
|
422 |
"step": 290
|
423 |
},
|
424 |
{
|
425 |
"epoch": 0.30857740585774057,
|
426 |
-
"grad_norm": 0.
|
427 |
"learning_rate": 1.7472049334512354e-05,
|
428 |
-
"loss": 1.
|
429 |
"step": 295
|
430 |
},
|
431 |
{
|
432 |
"epoch": 0.3138075313807531,
|
433 |
-
"grad_norm": 0.
|
434 |
"learning_rate": 1.7349421204029343e-05,
|
435 |
"loss": 1.0838,
|
436 |
"step": 300
|
437 |
},
|
438 |
{
|
439 |
"epoch": 0.3190376569037657,
|
440 |
-
"grad_norm": 0.
|
441 |
"learning_rate": 1.7224341279897473e-05,
|
442 |
"loss": 1.0937,
|
443 |
"step": 305
|
444 |
},
|
445 |
{
|
446 |
"epoch": 0.32426778242677823,
|
447 |
-
"grad_norm": 0.
|
448 |
"learning_rate": 1.7096851289234448e-05,
|
449 |
"loss": 1.0861,
|
450 |
"step": 310
|
451 |
},
|
452 |
{
|
453 |
"epoch": 0.3294979079497908,
|
454 |
-
"grad_norm": 0.
|
455 |
"learning_rate": 1.6966993763164928e-05,
|
456 |
"loss": 1.0863,
|
457 |
"step": 315
|
458 |
},
|
459 |
{
|
460 |
"epoch": 0.33472803347280333,
|
461 |
-
"grad_norm": 0.
|
462 |
"learning_rate": 1.6834812022632e-05,
|
463 |
-
"loss": 1.
|
464 |
"step": 320
|
465 |
},
|
466 |
{
|
467 |
"epoch": 0.3399581589958159,
|
468 |
-
"grad_norm": 0.
|
469 |
"learning_rate": 1.670035016394513e-05,
|
470 |
"loss": 1.0875,
|
471 |
"step": 325
|
472 |
},
|
473 |
{
|
474 |
"epoch": 0.34518828451882844,
|
475 |
-
"grad_norm": 0.
|
476 |
"learning_rate": 1.656365304406953e-05,
|
477 |
-
"loss": 1.
|
478 |
"step": 330
|
479 |
},
|
480 |
{
|
481 |
"epoch": 0.350418410041841,
|
482 |
-
"grad_norm": 0.
|
483 |
"learning_rate": 1.6424766265661648e-05,
|
484 |
"loss": 1.084,
|
485 |
"step": 335
|
486 |
},
|
487 |
{
|
488 |
"epoch": 0.35564853556485354,
|
489 |
-
"grad_norm": 0.
|
490 |
"learning_rate": 1.6283736161855995e-05,
|
491 |
"loss": 1.0772,
|
492 |
"step": 340
|
493 |
},
|
494 |
{
|
495 |
"epoch": 0.3608786610878661,
|
496 |
-
"grad_norm": 0.
|
497 |
"learning_rate": 1.614060978080823e-05,
|
498 |
-
"loss": 1.
|
499 |
"step": 345
|
500 |
},
|
501 |
{
|
502 |
"epoch": 0.36610878661087864,
|
503 |
-
"grad_norm": 0.
|
504 |
"learning_rate": 1.5995434869999723e-05,
|
505 |
"loss": 1.079,
|
506 |
"step": 350
|
507 |
},
|
508 |
{
|
509 |
"epoch": 0.3713389121338912,
|
510 |
-
"grad_norm": 0.
|
511 |
"learning_rate": 1.584825986030883e-05,
|
512 |
"loss": 1.0876,
|
513 |
"step": 355
|
514 |
},
|
515 |
{
|
516 |
"epoch": 0.37656903765690375,
|
517 |
-
"grad_norm": 0.
|
518 |
"learning_rate": 1.5699133849854164e-05,
|
519 |
-
"loss": 1.
|
520 |
"step": 360
|
521 |
},
|
522 |
{
|
523 |
"epoch": 0.3817991631799163,
|
524 |
-
"grad_norm": 0.
|
525 |
"learning_rate": 1.5548106587615295e-05,
|
526 |
-
"loss": 1.
|
527 |
"step": 365
|
528 |
},
|
529 |
{
|
530 |
"epoch": 0.38702928870292885,
|
531 |
-
"grad_norm": 0.
|
532 |
"learning_rate": 1.5395228456836298e-05,
|
533 |
"loss": 1.084,
|
534 |
"step": 370
|
535 |
},
|
536 |
{
|
537 |
"epoch": 0.3922594142259414,
|
538 |
-
"grad_norm": 0.
|
539 |
"learning_rate": 1.524055045821773e-05,
|
540 |
"loss": 1.0915,
|
541 |
"step": 375
|
542 |
},
|
543 |
{
|
544 |
"epoch": 0.39748953974895396,
|
545 |
-
"grad_norm": 0.
|
546 |
"learning_rate": 1.5084124192902612e-05,
|
547 |
"loss": 1.0788,
|
548 |
"step": 380
|
549 |
},
|
550 |
{
|
551 |
"epoch": 0.4027196652719665,
|
552 |
-
"grad_norm": 0.
|
553 |
"learning_rate": 1.4926001845262081e-05,
|
554 |
"loss": 1.0831,
|
555 |
"step": 385
|
556 |
},
|
557 |
{
|
558 |
"epoch": 0.40794979079497906,
|
559 |
-
"grad_norm": 0.
|
560 |
"learning_rate": 1.4766236165486526e-05,
|
561 |
"loss": 1.0798,
|
562 |
"step": 390
|
563 |
},
|
564 |
{
|
565 |
"epoch": 0.4131799163179916,
|
566 |
-
"grad_norm": 0.
|
567 |
"learning_rate": 1.4604880451987895e-05,
|
568 |
"loss": 1.0763,
|
569 |
"step": 395
|
570 |
},
|
571 |
{
|
572 |
"epoch": 0.41841004184100417,
|
573 |
-
"grad_norm": 0.
|
574 |
"learning_rate": 1.4441988533619182e-05,
|
575 |
"loss": 1.0735,
|
576 |
"step": 400
|
577 |
},
|
578 |
{
|
579 |
"epoch": 0.4236401673640167,
|
580 |
-
"grad_norm": 0.
|
581 |
"learning_rate": 1.4277614751716911e-05,
|
582 |
"loss": 1.0821,
|
583 |
"step": 405
|
584 |
},
|
585 |
{
|
586 |
"epoch": 0.42887029288702927,
|
587 |
-
"grad_norm": 0.
|
588 |
"learning_rate": 1.4111813941972672e-05,
|
589 |
"loss": 1.0961,
|
590 |
"step": 410
|
591 |
},
|
592 |
{
|
593 |
"epoch": 0.4341004184100418,
|
594 |
-
"grad_norm": 0.
|
595 |
"learning_rate": 1.3944641416139737e-05,
|
596 |
-
"loss": 1.
|
597 |
"step": 415
|
598 |
},
|
599 |
{
|
600 |
"epoch": 0.4393305439330544,
|
601 |
-
"grad_norm": 0.
|
602 |
"learning_rate": 1.3776152943580846e-05,
|
603 |
-
"loss": 1.
|
604 |
"step": 420
|
605 |
},
|
606 |
{
|
607 |
"epoch": 0.4445606694560669,
|
608 |
-
"grad_norm": 0.
|
609 |
"learning_rate": 1.3606404732663365e-05,
|
610 |
-
"loss": 1.
|
611 |
"step": 425
|
612 |
},
|
613 |
{
|
614 |
"epoch": 0.4497907949790795,
|
615 |
-
"grad_norm": 0.
|
616 |
"learning_rate": 1.3435453412007949e-05,
|
617 |
-
"loss": 1.
|
618 |
"step": 430
|
619 |
},
|
620 |
{
|
621 |
"epoch": 0.45502092050209203,
|
622 |
-
"grad_norm": 0.
|
623 |
"learning_rate": 1.3263356011597045e-05,
|
624 |
-
"loss": 1.
|
625 |
"step": 435
|
626 |
},
|
627 |
{
|
628 |
"epoch": 0.4602510460251046,
|
629 |
-
"grad_norm": 0.
|
630 |
"learning_rate": 1.3090169943749475e-05,
|
631 |
-
"loss": 1.
|
632 |
"step": 440
|
633 |
},
|
634 |
{
|
635 |
"epoch": 0.46548117154811713,
|
636 |
-
"grad_norm": 0.
|
637 |
"learning_rate": 1.2915952983967477e-05,
|
638 |
-
"loss": 1.
|
639 |
"step": 445
|
640 |
},
|
641 |
{
|
642 |
"epoch": 0.4707112970711297,
|
643 |
-
"grad_norm": 0.
|
644 |
"learning_rate": 1.2740763251662585e-05,
|
645 |
-
"loss": 1.
|
646 |
"step": 450
|
647 |
},
|
648 |
{
|
649 |
"epoch": 0.47594142259414224,
|
650 |
-
"grad_norm": 0.
|
651 |
"learning_rate": 1.2564659190766788e-05,
|
652 |
"loss": 1.0764,
|
653 |
"step": 455
|
654 |
},
|
655 |
{
|
656 |
"epoch": 0.4811715481171548,
|
657 |
-
"grad_norm": 0.
|
658 |
"learning_rate": 1.2387699550235419e-05,
|
659 |
"loss": 1.0858,
|
660 |
"step": 460
|
661 |
},
|
662 |
{
|
663 |
"epoch": 0.48640167364016734,
|
664 |
-
"grad_norm": 0.
|
665 |
"learning_rate": 1.2209943364448298e-05,
|
666 |
"loss": 1.0584,
|
667 |
"step": 465
|
668 |
},
|
669 |
{
|
670 |
"epoch": 0.4916317991631799,
|
671 |
-
"grad_norm": 0.
|
672 |
"learning_rate": 1.2031449933515625e-05,
|
673 |
"loss": 1.0687,
|
674 |
"step": 470
|
675 |
},
|
676 |
{
|
677 |
"epoch": 0.49686192468619245,
|
678 |
-
"grad_norm": 0.
|
679 |
"learning_rate": 1.1852278803495267e-05,
|
680 |
-
"loss": 1.
|
681 |
"step": 475
|
682 |
},
|
683 |
{
|
684 |
"epoch": 0.502092050209205,
|
685 |
-
"grad_norm": 0.
|
686 |
"learning_rate": 1.1672489746527979e-05,
|
687 |
"loss": 1.0758,
|
688 |
"step": 480
|
689 |
},
|
690 |
{
|
691 |
"epoch": 0.5073221757322176,
|
692 |
-
"grad_norm": 0.
|
693 |
"learning_rate": 1.1492142740897188e-05,
|
694 |
"loss": 1.0677,
|
695 |
"step": 485
|
696 |
},
|
697 |
{
|
698 |
"epoch": 0.5125523012552301,
|
699 |
-
"grad_norm": 0.
|
700 |
"learning_rate": 1.1311297951020028e-05,
|
701 |
"loss": 1.0652,
|
702 |
"step": 490
|
703 |
},
|
704 |
{
|
705 |
"epoch": 0.5177824267782427,
|
706 |
-
"grad_norm": 0.
|
707 |
"learning_rate": 1.1130015707376262e-05,
|
708 |
"loss": 1.0715,
|
709 |
"step": 495
|
710 |
},
|
711 |
{
|
712 |
"epoch": 0.5230125523012552,
|
713 |
-
"grad_norm": 0.
|
714 |
"learning_rate": 1.0948356486381829e-05,
|
715 |
"loss": 1.0769,
|
716 |
"step": 500
|
717 |
},
|
718 |
{
|
719 |
"epoch": 0.5282426778242678,
|
720 |
-
"grad_norm": 0.
|
721 |
"learning_rate": 1.0766380890213675e-05,
|
722 |
"loss": 1.0806,
|
723 |
"step": 505
|
724 |
},
|
725 |
{
|
726 |
"epoch": 0.5334728033472803,
|
727 |
-
"grad_norm": 0.
|
728 |
"learning_rate": 1.0584149626592662e-05,
|
729 |
"loss": 1.056,
|
730 |
"step": 510
|
731 |
},
|
732 |
{
|
733 |
"epoch": 0.5387029288702929,
|
734 |
-
"grad_norm": 0.
|
735 |
"learning_rate": 1.0401723488531238e-05,
|
736 |
-
"loss": 1.
|
737 |
"step": 515
|
738 |
},
|
739 |
{
|
740 |
"epoch": 0.5439330543933054,
|
741 |
-
"grad_norm": 0.
|
742 |
"learning_rate": 1.0219163334052682e-05,
|
743 |
-
"loss": 1.
|
744 |
"step": 520
|
745 |
},
|
746 |
{
|
747 |
"epoch": 0.549163179916318,
|
748 |
-
"grad_norm": 0.
|
749 |
"learning_rate": 1.003653006588863e-05,
|
750 |
"loss": 1.0757,
|
751 |
"step": 525
|
752 |
},
|
753 |
{
|
754 |
"epoch": 0.5543933054393305,
|
755 |
-
"grad_norm": 0.
|
756 |
"learning_rate": 9.853884611161709e-06,
|
757 |
"loss": 1.047,
|
758 |
"step": 530
|
759 |
},
|
760 |
{
|
761 |
"epoch": 0.5596234309623431,
|
762 |
-
"grad_norm": 0.
|
763 |
"learning_rate": 9.671287901060035e-06,
|
764 |
"loss": 1.064,
|
765 |
"step": 535
|
766 |
},
|
767 |
{
|
768 |
"epoch": 0.5648535564853556,
|
769 |
-
"grad_norm": 0.
|
770 |
"learning_rate": 9.48880085051033e-06,
|
771 |
"loss": 1.0734,
|
772 |
"step": 540
|
773 |
},
|
774 |
{
|
775 |
"epoch": 0.5700836820083682,
|
776 |
-
"grad_norm": 0.
|
777 |
"learning_rate": 9.306484337856497e-06,
|
778 |
-
"loss": 1.
|
779 |
"step": 545
|
780 |
},
|
781 |
{
|
782 |
"epoch": 0.5753138075313807,
|
783 |
-
"grad_norm": 0.
|
784 |
"learning_rate": 9.124399184550377e-06,
|
785 |
-
"loss": 1.
|
786 |
"step": 550
|
787 |
},
|
788 |
{
|
789 |
"epoch": 0.5805439330543933,
|
790 |
-
"grad_norm": 0.
|
791 |
"learning_rate": 8.94260613486149e-06,
|
792 |
"loss": 1.0757,
|
793 |
"step": 555
|
794 |
},
|
795 |
{
|
796 |
"epoch": 0.5857740585774058,
|
797 |
-
"grad_norm": 0.
|
798 |
"learning_rate": 8.76116583561252e-06,
|
799 |
-
"loss": 1.
|
800 |
"step": 560
|
801 |
},
|
802 |
{
|
803 |
"epoch": 0.5910041841004184,
|
804 |
-
"grad_norm": 0.
|
805 |
"learning_rate": 8.580138815947336e-06,
|
806 |
"loss": 1.0651,
|
807 |
"step": 565
|
808 |
},
|
809 |
{
|
810 |
"epoch": 0.5962343096234309,
|
811 |
-
"grad_norm": 0.
|
812 |
"learning_rate": 8.399585467138215e-06,
|
813 |
"loss": 1.0763,
|
814 |
"step": 570
|
815 |
},
|
816 |
{
|
817 |
"epoch": 0.6014644351464435,
|
818 |
-
"grad_norm": 0.
|
819 |
"learning_rate": 8.219566022439128e-06,
|
820 |
"loss": 1.0792,
|
821 |
"step": 575
|
822 |
},
|
823 |
{
|
824 |
"epoch": 0.606694560669456,
|
825 |
-
"grad_norm": 0.
|
826 |
"learning_rate": 8.040140536991688e-06,
|
827 |
-
"loss": 1.
|
828 |
"step": 580
|
829 |
},
|
830 |
{
|
831 |
"epoch": 0.6119246861924686,
|
832 |
-
"grad_norm": 0.
|
833 |
"learning_rate": 7.861368867790541e-06,
|
834 |
-
"loss": 1.
|
835 |
"step": 585
|
836 |
},
|
837 |
{
|
838 |
"epoch": 0.6171548117154811,
|
839 |
-
"grad_norm": 0.
|
840 |
"learning_rate": 7.683310653714857e-06,
|
841 |
"loss": 1.0723,
|
842 |
"step": 590
|
843 |
},
|
844 |
{
|
845 |
"epoch": 0.6223849372384938,
|
846 |
-
"grad_norm": 0.
|
847 |
"learning_rate": 7.506025295632577e-06,
|
848 |
-
"loss": 1.
|
849 |
"step": 595
|
850 |
},
|
851 |
{
|
852 |
"epoch": 0.6276150627615062,
|
853 |
-
"grad_norm": 0.
|
854 |
"learning_rate": 7.329571936584072e-06,
|
855 |
"loss": 1.0566,
|
856 |
"step": 600
|
857 |
},
|
858 |
{
|
859 |
"epoch": 0.6328451882845189,
|
860 |
-
"grad_norm": 0.
|
861 |
"learning_rate": 7.154009442051813e-06,
|
862 |
"loss": 1.0562,
|
863 |
"step": 605
|
864 |
},
|
865 |
{
|
866 |
"epoch": 0.6380753138075314,
|
867 |
-
"grad_norm": 0.
|
868 |
"learning_rate": 6.979396380322621e-06,
|
869 |
"loss": 1.0744,
|
870 |
"step": 610
|
871 |
},
|
872 |
{
|
873 |
"epoch": 0.643305439330544,
|
874 |
-
"grad_norm": 0.
|
875 |
"learning_rate": 6.805791002949085e-06,
|
876 |
"loss": 1.0688,
|
877 |
"step": 615
|
878 |
},
|
879 |
{
|
880 |
"epoch": 0.6485355648535565,
|
881 |
-
"grad_norm": 0.
|
882 |
"learning_rate": 6.63325122531663e-06,
|
883 |
"loss": 1.0541,
|
884 |
"step": 620
|
885 |
},
|
886 |
{
|
887 |
"epoch": 0.6537656903765691,
|
888 |
-
"grad_norm": 0.
|
889 |
"learning_rate": 6.461834607322719e-06,
|
890 |
"loss": 1.0547,
|
891 |
"step": 625
|
892 |
},
|
893 |
{
|
894 |
"epoch": 0.6589958158995816,
|
895 |
-
"grad_norm": 0.
|
896 |
"learning_rate": 6.291598334174685e-06,
|
897 |
"loss": 1.0572,
|
898 |
"step": 630
|
899 |
},
|
900 |
{
|
901 |
"epoch": 0.6642259414225942,
|
902 |
-
"grad_norm": 0.
|
903 |
"learning_rate": 6.122599197312525e-06,
|
904 |
"loss": 1.0858,
|
905 |
"step": 635
|
906 |
},
|
907 |
{
|
908 |
"epoch": 0.6694560669456067,
|
909 |
-
"grad_norm": 0.
|
910 |
"learning_rate": 5.954893575463064e-06,
|
911 |
"loss": 1.0637,
|
912 |
"step": 640
|
913 |
},
|
914 |
{
|
915 |
"epoch": 0.6746861924686193,
|
916 |
-
"grad_norm": 0.
|
917 |
"learning_rate": 5.788537415831825e-06,
|
918 |
"loss": 1.055,
|
919 |
"step": 645
|
920 |
},
|
921 |
{
|
922 |
"epoch": 0.6799163179916318,
|
923 |
-
"grad_norm": 0.
|
924 |
"learning_rate": 5.623586215438813e-06,
|
925 |
"loss": 1.0641,
|
926 |
"step": 650
|
927 |
},
|
928 |
{
|
929 |
"epoch": 0.6851464435146444,
|
930 |
-
"grad_norm": 0.
|
931 |
"learning_rate": 5.460095002604533e-06,
|
932 |
"loss": 1.0477,
|
933 |
"step": 655
|
934 |
},
|
935 |
{
|
936 |
"epoch": 0.6903765690376569,
|
937 |
-
"grad_norm": 0.
|
938 |
"learning_rate": 5.298118318592316e-06,
|
939 |
"loss": 1.0654,
|
940 |
"step": 660
|
941 |
},
|
942 |
{
|
943 |
"epoch": 0.6956066945606695,
|
944 |
-
"grad_norm": 0.
|
945 |
"learning_rate": 5.137710199413182e-06,
|
946 |
-
"loss": 1.
|
947 |
"step": 665
|
948 |
},
|
949 |
{
|
950 |
"epoch": 0.700836820083682,
|
951 |
-
"grad_norm": 0.
|
952 |
"learning_rate": 4.978924157799208e-06,
|
953 |
-
"loss": 1.
|
954 |
"step": 670
|
955 |
},
|
956 |
{
|
957 |
"epoch": 0.7060669456066946,
|
958 |
-
"grad_norm": 0.
|
959 |
"learning_rate": 4.821813165351495e-06,
|
960 |
"loss": 1.0581,
|
961 |
"step": 675
|
962 |
},
|
963 |
{
|
964 |
"epoch": 0.7112970711297071,
|
965 |
-
"grad_norm": 0.
|
966 |
"learning_rate": 4.666429634868651e-06,
|
967 |
"loss": 1.0851,
|
968 |
"step": 680
|
969 |
},
|
970 |
{
|
971 |
"epoch": 0.7165271966527197,
|
972 |
-
"grad_norm": 0.
|
973 |
"learning_rate": 4.512825402861686e-06,
|
974 |
"loss": 1.0552,
|
975 |
"step": 685
|
976 |
},
|
977 |
{
|
978 |
"epoch": 0.7217573221757322,
|
979 |
-
"grad_norm": 0.
|
980 |
"learning_rate": 4.361051712261173e-06,
|
981 |
"loss": 1.0596,
|
982 |
"step": 690
|
983 |
},
|
984 |
{
|
985 |
"epoch": 0.7269874476987448,
|
986 |
-
"grad_norm": 0.
|
987 |
"learning_rate": 4.211159195322404e-06,
|
988 |
-
"loss": 1.
|
989 |
"step": 695
|
990 |
},
|
991 |
{
|
992 |
"epoch": 0.7322175732217573,
|
993 |
-
"grad_norm": 0.
|
994 |
"learning_rate": 4.063197856734295e-06,
|
995 |
"loss": 1.0783,
|
996 |
"step": 700
|
997 |
},
|
998 |
{
|
999 |
"epoch": 0.7374476987447699,
|
1000 |
-
"grad_norm": 0.
|
1001 |
"learning_rate": 3.917217056937649e-06,
|
1002 |
"loss": 1.0636,
|
1003 |
"step": 705
|
1004 |
},
|
1005 |
{
|
1006 |
"epoch": 0.7426778242677824,
|
1007 |
-
"grad_norm": 0.
|
1008 |
"learning_rate": 3.773265495658309e-06,
|
1009 |
"loss": 1.0655,
|
1010 |
"step": 710
|
1011 |
},
|
1012 |
{
|
1013 |
"epoch": 0.747907949790795,
|
1014 |
-
"grad_norm": 0.
|
1015 |
"learning_rate": 3.6313911956607827e-06,
|
1016 |
-
"loss": 1.
|
1017 |
"step": 715
|
1018 |
},
|
1019 |
{
|
1020 |
"epoch": 0.7531380753138075,
|
1021 |
-
"grad_norm": 0.
|
1022 |
"learning_rate": 3.491641486727645e-06,
|
1023 |
"loss": 1.0684,
|
1024 |
"step": 720
|
1025 |
},
|
1026 |
{
|
1027 |
"epoch": 0.7583682008368201,
|
1028 |
-
"grad_norm": 0.
|
1029 |
"learning_rate": 3.354062989870166e-06,
|
1030 |
-
"loss": 1.
|
1031 |
"step": 725
|
1032 |
},
|
1033 |
{
|
1034 |
"epoch": 0.7635983263598326,
|
1035 |
-
"grad_norm": 0.
|
1036 |
"learning_rate": 3.2187016017753714e-06,
|
1037 |
"loss": 1.066,
|
1038 |
"step": 730
|
1039 |
},
|
1040 |
{
|
1041 |
"epoch": 0.7688284518828452,
|
1042 |
-
"grad_norm": 0.
|
1043 |
"learning_rate": 3.085602479494735e-06,
|
1044 |
"loss": 1.0602,
|
1045 |
"step": 735
|
1046 |
},
|
1047 |
{
|
1048 |
"epoch": 0.7740585774058577,
|
1049 |
-
"grad_norm": 0.
|
1050 |
"learning_rate": 2.954810025379633e-06,
|
1051 |
"loss": 1.071,
|
1052 |
"step": 740
|
1053 |
},
|
1054 |
{
|
1055 |
"epoch": 0.7792887029288703,
|
1056 |
-
"grad_norm": 0.
|
1057 |
"learning_rate": 2.826367872268543e-06,
|
1058 |
"loss": 1.0517,
|
1059 |
"step": 745
|
1060 |
},
|
1061 |
{
|
1062 |
"epoch": 0.7845188284518828,
|
1063 |
-
"grad_norm": 0.
|
1064 |
"learning_rate": 2.700318868930977e-06,
|
1065 |
"loss": 1.0683,
|
1066 |
"step": 750
|
1067 |
},
|
1068 |
{
|
1069 |
"epoch": 0.7897489539748954,
|
1070 |
-
"grad_norm": 0.
|
1071 |
"learning_rate": 2.576705065772985e-06,
|
1072 |
"loss": 1.0543,
|
1073 |
"step": 755
|
1074 |
},
|
1075 |
{
|
1076 |
"epoch": 0.7949790794979079,
|
1077 |
-
"grad_norm": 0.
|
1078 |
"learning_rate": 2.455567700808974e-06,
|
1079 |
"loss": 1.0437,
|
1080 |
"step": 760
|
1081 |
},
|
1082 |
{
|
1083 |
"epoch": 0.8002092050209205,
|
1084 |
-
"grad_norm": 0.
|
1085 |
"learning_rate": 2.3369471859045558e-06,
|
1086 |
"loss": 1.0637,
|
1087 |
"step": 765
|
1088 |
},
|
1089 |
{
|
1090 |
"epoch": 0.805439330543933,
|
1091 |
-
"grad_norm": 0.
|
1092 |
"learning_rate": 2.2208830932950175e-06,
|
1093 |
-
"loss": 1.
|
1094 |
"step": 770
|
1095 |
},
|
1096 |
{
|
1097 |
"epoch": 0.8106694560669456,
|
1098 |
-
"grad_norm": 0.
|
1099 |
"learning_rate": 2.107414142383857e-06,
|
1100 |
"loss": 1.0636,
|
1101 |
"step": 775
|
1102 |
},
|
1103 |
{
|
1104 |
"epoch": 0.8158995815899581,
|
1105 |
-
"grad_norm": 0.
|
1106 |
"learning_rate": 1.996578186825876e-06,
|
1107 |
"loss": 1.0652,
|
1108 |
"step": 780
|
1109 |
},
|
1110 |
{
|
1111 |
"epoch": 0.8211297071129707,
|
1112 |
-
"grad_norm": 0.
|
1113 |
"learning_rate": 1.8884122018990292e-06,
|
1114 |
-
"loss": 1.
|
1115 |
"step": 785
|
1116 |
},
|
1117 |
{
|
1118 |
"epoch": 0.8263598326359832,
|
1119 |
-
"grad_norm": 0.
|
1120 |
"learning_rate": 1.7829522721693738e-06,
|
1121 |
"loss": 1.0476,
|
1122 |
"step": 790
|
1123 |
},
|
1124 |
{
|
1125 |
"epoch": 0.8315899581589958,
|
1126 |
-
"grad_norm": 0.
|
1127 |
"learning_rate": 1.680233579453101e-06,
|
1128 |
"loss": 1.0604,
|
1129 |
"step": 795
|
1130 |
},
|
1131 |
{
|
1132 |
"epoch": 0.8368200836820083,
|
1133 |
-
"grad_norm": 0.
|
1134 |
"learning_rate": 1.5802903910797584e-06,
|
1135 |
"loss": 1.0623,
|
1136 |
"step": 800
|
1137 |
},
|
1138 |
{
|
1139 |
"epoch": 0.8420502092050209,
|
1140 |
-
"grad_norm": 0.
|
1141 |
"learning_rate": 1.483156048460561e-06,
|
1142 |
"loss": 1.0702,
|
1143 |
"step": 805
|
1144 |
},
|
1145 |
{
|
1146 |
"epoch": 0.8472803347280334,
|
1147 |
-
"grad_norm": 0.
|
1148 |
"learning_rate": 1.3888629559655497e-06,
|
1149 |
"loss": 1.065,
|
1150 |
"step": 810
|
1151 |
},
|
1152 |
{
|
1153 |
"epoch": 0.852510460251046,
|
1154 |
-
"grad_norm": 0.
|
1155 |
"learning_rate": 1.2974425701133941e-06,
|
1156 |
"loss": 1.0519,
|
1157 |
"step": 815
|
1158 |
},
|
1159 |
{
|
1160 |
"epoch": 0.8577405857740585,
|
1161 |
-
"grad_norm": 0.
|
1162 |
"learning_rate": 1.2089253890773789e-06,
|
1163 |
"loss": 1.0592,
|
1164 |
"step": 820
|
1165 |
},
|
1166 |
{
|
1167 |
"epoch": 0.8629707112970711,
|
1168 |
-
"grad_norm": 0.
|
1169 |
"learning_rate": 1.1233409425110952e-06,
|
1170 |
"loss": 1.0562,
|
1171 |
"step": 825
|
1172 |
},
|
1173 |
{
|
1174 |
"epoch": 0.8682008368200836,
|
1175 |
-
"grad_norm": 0.
|
1176 |
"learning_rate": 1.0407177816972558e-06,
|
1177 |
"loss": 1.0469,
|
1178 |
"step": 830
|
1179 |
},
|
1180 |
{
|
1181 |
"epoch": 0.8734309623430963,
|
1182 |
-
"grad_norm": 0.
|
1183 |
"learning_rate": 9.610834700228654e-07,
|
1184 |
-
"loss": 1.
|
1185 |
"step": 835
|
1186 |
},
|
1187 |
{
|
1188 |
"epoch": 0.8786610878661087,
|
1189 |
-
"grad_norm": 0.
|
1190 |
"learning_rate": 8.844645737839874e-07,
|
1191 |
-
"loss": 1.
|
1192 |
"step": 840
|
1193 |
},
|
1194 |
{
|
1195 |
"epoch": 0.8838912133891214,
|
1196 |
-
"grad_norm": 0.
|
1197 |
"learning_rate": 8.108866533231252e-07,
|
1198 |
-
"loss": 1.
|
1199 |
"step": 845
|
1200 |
},
|
1201 |
{
|
1202 |
"epoch": 0.8891213389121339,
|
1203 |
-
"grad_norm": 0.
|
1204 |
"learning_rate": 7.403742545021986e-07,
|
1205 |
"loss": 1.0559,
|
1206 |
"step": 850
|
1207 |
},
|
1208 |
{
|
1209 |
"epoch": 0.8943514644351465,
|
1210 |
-
"grad_norm": 0.
|
1211 |
"learning_rate": 6.729509005139634e-07,
|
1212 |
"loss": 1.0415,
|
1213 |
"step": 855
|
1214 |
},
|
1215 |
{
|
1216 |
"epoch": 0.899581589958159,
|
1217 |
-
"grad_norm": 0.
|
1218 |
"learning_rate": 6.086390840345758e-07,
|
1219 |
-
"loss": 1.
|
1220 |
"step": 860
|
1221 |
},
|
1222 |
{
|
1223 |
"epoch": 0.9048117154811716,
|
1224 |
-
"grad_norm": 0.
|
1225 |
"learning_rate": 5.474602597199607e-07,
|
1226 |
"loss": 1.0704,
|
1227 |
"step": 865
|
1228 |
},
|
1229 |
{
|
1230 |
"epoch": 0.9100418410041841,
|
1231 |
-
"grad_norm": 0.
|
1232 |
"learning_rate": 4.894348370484648e-07,
|
1233 |
"loss": 1.0691,
|
1234 |
"step": 870
|
1235 |
},
|
1236 |
{
|
1237 |
"epoch": 0.9152719665271967,
|
1238 |
-
"grad_norm": 0.
|
1239 |
"learning_rate": 4.34582173512168e-07,
|
1240 |
-
"loss": 1.
|
1241 |
"step": 875
|
1242 |
},
|
1243 |
{
|
1244 |
"epoch": 0.9205020920502092,
|
1245 |
-
"grad_norm": 0.
|
1246 |
"learning_rate": 3.8292056815916965e-07,
|
1247 |
"loss": 1.059,
|
1248 |
"step": 880
|
1249 |
},
|
1250 |
{
|
1251 |
"epoch": 0.9257322175732218,
|
1252 |
-
"grad_norm": 0.
|
1253 |
"learning_rate": 3.344672554889428e-07,
|
1254 |
-
"loss": 1.
|
1255 |
"step": 885
|
1256 |
},
|
1257 |
{
|
1258 |
"epoch": 0.9309623430962343,
|
1259 |
-
"grad_norm": 0.
|
1260 |
"learning_rate": 2.8923839970285473e-07,
|
1261 |
"loss": 1.0592,
|
1262 |
"step": 890
|
1263 |
},
|
1264 |
{
|
1265 |
"epoch": 0.9361924686192469,
|
1266 |
-
"grad_norm": 0.
|
1267 |
"learning_rate": 2.472490893117274e-07,
|
1268 |
"loss": 1.0458,
|
1269 |
"step": 895
|
1270 |
},
|
1271 |
{
|
1272 |
"epoch": 0.9414225941422594,
|
1273 |
-
"grad_norm": 0.
|
1274 |
"learning_rate": 2.0851333210225032e-07,
|
1275 |
"loss": 1.0608,
|
1276 |
"step": 900
|
1277 |
},
|
1278 |
{
|
1279 |
"epoch": 0.946652719665272,
|
1280 |
-
"grad_norm": 0.
|
1281 |
"learning_rate": 1.730440504639408e-07,
|
1282 |
-
"loss": 1.
|
1283 |
"step": 905
|
1284 |
},
|
1285 |
{
|
1286 |
"epoch": 0.9518828451882845,
|
1287 |
-
"grad_norm": 0.
|
1288 |
"learning_rate": 1.408530770781813e-07,
|
1289 |
-
"loss": 1.
|
1290 |
"step": 910
|
1291 |
},
|
1292 |
{
|
1293 |
"epoch": 0.9571129707112971,
|
1294 |
-
"grad_norm": 0.
|
1295 |
"learning_rate": 1.1195115097079268e-07,
|
1296 |
-
"loss": 1.
|
1297 |
"step": 915
|
1298 |
},
|
1299 |
{
|
1300 |
"epoch": 0.9623430962343096,
|
1301 |
-
"grad_norm": 0.
|
1302 |
"learning_rate": 8.634791392946429e-08,
|
1303 |
-
"loss": 1.
|
1304 |
"step": 920
|
1305 |
},
|
1306 |
{
|
1307 |
"epoch": 0.9675732217573222,
|
1308 |
-
"grad_norm": 0.
|
1309 |
"learning_rate": 6.405190728721033e-08,
|
1310 |
"loss": 1.0455,
|
1311 |
"step": 925
|
1312 |
},
|
1313 |
{
|
1314 |
"epoch": 0.9728033472803347,
|
1315 |
-
"grad_norm": 0.
|
1316 |
"learning_rate": 4.5070569072952485e-08,
|
1317 |
-
"loss": 1.
|
1318 |
"step": 930
|
1319 |
},
|
1320 |
{
|
1321 |
"epoch": 0.9780334728033473,
|
1322 |
-
"grad_norm": 0.
|
1323 |
"learning_rate": 2.9410231530168087e-08,
|
1324 |
-
"loss": 1.
|
1325 |
"step": 935
|
1326 |
},
|
1327 |
{
|
1328 |
"epoch": 0.9832635983263598,
|
1329 |
-
"grad_norm": 0.
|
1330 |
"learning_rate": 1.7076119004429958e-08,
|
1331 |
"loss": 1.0763,
|
1332 |
"step": 940
|
1333 |
},
|
1334 |
{
|
1335 |
"epoch": 0.9884937238493724,
|
1336 |
-
"grad_norm": 0.
|
1337 |
"learning_rate": 8.072346200544979e-09,
|
1338 |
"loss": 1.0672,
|
1339 |
"step": 945
|
1340 |
},
|
1341 |
{
|
1342 |
"epoch": 0.9937238493723849,
|
1343 |
-
"grad_norm": 0.
|
1344 |
"learning_rate": 2.401916809872118e-09,
|
1345 |
-
"loss": 1.
|
1346 |
"step": 950
|
1347 |
},
|
1348 |
{
|
1349 |
"epoch": 0.9989539748953975,
|
1350 |
-
"grad_norm": 0.
|
1351 |
"learning_rate": 6.672250828620996e-11,
|
1352 |
"loss": 1.0627,
|
1353 |
"step": 955
|
1354 |
},
|
1355 |
{
|
1356 |
"epoch": 1.0,
|
1357 |
-
"eval_loss": 1.
|
1358 |
-
"eval_runtime":
|
1359 |
-
"eval_samples_per_second": 47.
|
1360 |
-
"eval_steps_per_second": 0.
|
1361 |
"step": 956
|
1362 |
},
|
1363 |
{
|
1364 |
"epoch": 1.0,
|
1365 |
"step": 956,
|
1366 |
-
"total_flos":
|
1367 |
-
"train_loss":
|
1368 |
-
"train_runtime":
|
1369 |
-
"train_samples_per_second":
|
1370 |
-
"train_steps_per_second":
|
1371 |
}
|
1372 |
],
|
1373 |
"logging_steps": 5,
|
@@ -1375,7 +1375,7 @@
|
|
1375 |
"num_input_tokens_seen": 0,
|
1376 |
"num_train_epochs": 1,
|
1377 |
"save_steps": 100,
|
1378 |
-
"total_flos":
|
1379 |
"train_batch_size": 8,
|
1380 |
"trial_name": null,
|
1381 |
"trial_params": null
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.0010460251046025104,
|
13 |
+
"grad_norm": 4.858334608923654,
|
14 |
"learning_rate": 2.0833333333333333e-07,
|
15 |
"loss": 1.3092,
|
16 |
"step": 1
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.005230125523012552,
|
20 |
+
"grad_norm": 4.6974531972545845,
|
21 |
"learning_rate": 1.0416666666666667e-06,
|
22 |
"loss": 1.2901,
|
23 |
"step": 5
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.010460251046025104,
|
27 |
+
"grad_norm": 4.050379213449396,
|
28 |
"learning_rate": 2.0833333333333334e-06,
|
29 |
"loss": 1.2488,
|
30 |
"step": 10
|
31 |
},
|
32 |
{
|
33 |
"epoch": 0.015690376569037656,
|
34 |
+
"grad_norm": 1.260482829918926,
|
35 |
"learning_rate": 3.125e-06,
|
36 |
+
"loss": 1.1402,
|
37 |
"step": 15
|
38 |
},
|
39 |
{
|
40 |
"epoch": 0.02092050209205021,
|
41 |
+
"grad_norm": 0.7572315531712823,
|
42 |
"learning_rate": 4.166666666666667e-06,
|
43 |
"loss": 1.1238,
|
44 |
"step": 20
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.02615062761506276,
|
48 |
+
"grad_norm": 0.6345745967373482,
|
49 |
"learning_rate": 5.208333333333334e-06,
|
50 |
"loss": 1.0882,
|
51 |
"step": 25
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.03138075313807531,
|
55 |
+
"grad_norm": 0.5480254516819103,
|
56 |
"learning_rate": 6.25e-06,
|
57 |
"loss": 1.0972,
|
58 |
"step": 30
|
59 |
},
|
60 |
{
|
61 |
"epoch": 0.036610878661087864,
|
62 |
+
"grad_norm": 0.5577040974895372,
|
63 |
"learning_rate": 7.291666666666667e-06,
|
64 |
"loss": 1.0967,
|
65 |
"step": 35
|
66 |
},
|
67 |
{
|
68 |
"epoch": 0.04184100418410042,
|
69 |
+
"grad_norm": 0.5806448242568741,
|
70 |
"learning_rate": 8.333333333333334e-06,
|
71 |
"loss": 1.0806,
|
72 |
"step": 40
|
73 |
},
|
74 |
{
|
75 |
"epoch": 0.04707112970711297,
|
76 |
+
"grad_norm": 0.6520252066307457,
|
77 |
"learning_rate": 9.375000000000001e-06,
|
78 |
"loss": 1.0799,
|
79 |
"step": 45
|
80 |
},
|
81 |
{
|
82 |
"epoch": 0.05230125523012552,
|
83 |
+
"grad_norm": 0.5935927194187152,
|
84 |
"learning_rate": 1.0416666666666668e-05,
|
85 |
"loss": 1.0776,
|
86 |
"step": 50
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.05753138075313807,
|
90 |
+
"grad_norm": 0.539119622245633,
|
91 |
"learning_rate": 1.1458333333333333e-05,
|
92 |
"loss": 1.0899,
|
93 |
"step": 55
|
94 |
},
|
95 |
{
|
96 |
"epoch": 0.06276150627615062,
|
97 |
+
"grad_norm": 0.5602624433341231,
|
98 |
"learning_rate": 1.25e-05,
|
99 |
+
"loss": 1.0765,
|
100 |
"step": 60
|
101 |
},
|
102 |
{
|
103 |
"epoch": 0.06799163179916318,
|
104 |
+
"grad_norm": 0.6159503903121701,
|
105 |
"learning_rate": 1.3541666666666668e-05,
|
106 |
"loss": 1.0822,
|
107 |
"step": 65
|
108 |
},
|
109 |
{
|
110 |
"epoch": 0.07322175732217573,
|
111 |
+
"grad_norm": 0.6004379020843004,
|
112 |
"learning_rate": 1.4583333333333333e-05,
|
113 |
+
"loss": 1.0676,
|
114 |
"step": 70
|
115 |
},
|
116 |
{
|
117 |
"epoch": 0.07845188284518828,
|
118 |
+
"grad_norm": 0.5874082854599968,
|
119 |
"learning_rate": 1.5625e-05,
|
120 |
"loss": 1.094,
|
121 |
"step": 75
|
122 |
},
|
123 |
{
|
124 |
"epoch": 0.08368200836820083,
|
125 |
+
"grad_norm": 0.6751852785193894,
|
126 |
"learning_rate": 1.6666666666666667e-05,
|
127 |
"loss": 1.0929,
|
128 |
"step": 80
|
129 |
},
|
130 |
{
|
131 |
"epoch": 0.08891213389121339,
|
132 |
+
"grad_norm": 0.5828293222590728,
|
133 |
"learning_rate": 1.7708333333333335e-05,
|
134 |
"loss": 1.1021,
|
135 |
"step": 85
|
136 |
},
|
137 |
{
|
138 |
"epoch": 0.09414225941422594,
|
139 |
+
"grad_norm": 0.6900402033657895,
|
140 |
"learning_rate": 1.8750000000000002e-05,
|
141 |
"loss": 1.0856,
|
142 |
"step": 90
|
143 |
},
|
144 |
{
|
145 |
"epoch": 0.09937238493723849,
|
146 |
+
"grad_norm": 0.5853837848957884,
|
147 |
"learning_rate": 1.979166666666667e-05,
|
148 |
"loss": 1.0862,
|
149 |
"step": 95
|
150 |
},
|
151 |
{
|
152 |
"epoch": 0.10460251046025104,
|
153 |
+
"grad_norm": 0.6701191850895604,
|
154 |
"learning_rate": 1.9998932457674904e-05,
|
155 |
"loss": 1.0933,
|
156 |
"step": 100
|
157 |
},
|
158 |
{
|
159 |
"epoch": 0.1098326359832636,
|
160 |
+
"grad_norm": 0.5533200148308522,
|
161 |
"learning_rate": 1.999459595761682e-05,
|
162 |
"loss": 1.1147,
|
163 |
"step": 105
|
164 |
},
|
165 |
{
|
166 |
"epoch": 0.11506276150627615,
|
167 |
+
"grad_norm": 0.7369714526144382,
|
168 |
"learning_rate": 1.9986925223989665e-05,
|
169 |
"loss": 1.093,
|
170 |
"step": 110
|
171 |
},
|
172 |
{
|
173 |
"epoch": 0.1202928870292887,
|
174 |
+
"grad_norm": 0.5547288913187312,
|
175 |
"learning_rate": 1.9975922815778082e-05,
|
176 |
"loss": 1.0831,
|
177 |
"step": 115
|
178 |
},
|
179 |
{
|
180 |
"epoch": 0.12552301255230125,
|
181 |
+
"grad_norm": 0.6897888287129971,
|
182 |
"learning_rate": 1.996159240342547e-05,
|
183 |
"loss": 1.0825,
|
184 |
"step": 120
|
185 |
},
|
186 |
{
|
187 |
"epoch": 0.1307531380753138,
|
188 |
+
"grad_norm": 0.6130228169149287,
|
189 |
"learning_rate": 1.9943938767609515e-05,
|
190 |
+
"loss": 1.083,
|
191 |
"step": 125
|
192 |
},
|
193 |
{
|
194 |
"epoch": 0.13598326359832635,
|
195 |
+
"grad_norm": 0.5733760917524161,
|
196 |
"learning_rate": 1.9922967797647357e-05,
|
197 |
"loss": 1.0905,
|
198 |
"step": 130
|
199 |
},
|
200 |
{
|
201 |
"epoch": 0.1412133891213389,
|
202 |
+
"grad_norm": 0.5893547730249119,
|
203 |
"learning_rate": 1.9898686489530852e-05,
|
204 |
"loss": 1.0974,
|
205 |
"step": 135
|
206 |
},
|
207 |
{
|
208 |
"epoch": 0.14644351464435146,
|
209 |
+
"grad_norm": 0.5801179495005143,
|
210 |
"learning_rate": 1.9871102943592717e-05,
|
211 |
"loss": 1.0731,
|
212 |
"step": 140
|
213 |
},
|
214 |
{
|
215 |
"epoch": 0.151673640167364,
|
216 |
+
"grad_norm": 0.5638361223609211,
|
217 |
"learning_rate": 1.9840226361804223e-05,
|
218 |
"loss": 1.1055,
|
219 |
"step": 145
|
220 |
},
|
221 |
{
|
222 |
"epoch": 0.15690376569037656,
|
223 |
+
"grad_norm": 0.5449934219233721,
|
224 |
"learning_rate": 1.9806067044705375e-05,
|
225 |
+
"loss": 1.0761,
|
226 |
"step": 150
|
227 |
},
|
228 |
{
|
229 |
"epoch": 0.16213389121338911,
|
230 |
+
"grad_norm": 0.6274931137807601,
|
231 |
"learning_rate": 1.9768636387968616e-05,
|
232 |
"loss": 1.1064,
|
233 |
"step": 155
|
234 |
},
|
235 |
{
|
236 |
"epoch": 0.16736401673640167,
|
237 |
+
"grad_norm": 0.6250809185735244,
|
238 |
"learning_rate": 1.9727946878597193e-05,
|
239 |
"loss": 1.0791,
|
240 |
"step": 160
|
241 |
},
|
242 |
{
|
243 |
"epoch": 0.17259414225941422,
|
244 |
+
"grad_norm": 0.5284412080420968,
|
245 |
"learning_rate": 1.968401209075946e-05,
|
246 |
"loss": 1.0839,
|
247 |
"step": 165
|
248 |
},
|
249 |
{
|
250 |
"epoch": 0.17782426778242677,
|
251 |
+
"grad_norm": 0.5821294853027748,
|
252 |
"learning_rate": 1.963684668126046e-05,
|
253 |
+
"loss": 1.1018,
|
254 |
"step": 170
|
255 |
},
|
256 |
{
|
257 |
"epoch": 0.18305439330543932,
|
258 |
+
"grad_norm": 0.6298467979748181,
|
259 |
"learning_rate": 1.9586466384652366e-05,
|
260 |
"loss": 1.0847,
|
261 |
"step": 175
|
262 |
},
|
263 |
{
|
264 |
"epoch": 0.18828451882845187,
|
265 |
+
"grad_norm": 0.6400162596224541,
|
266 |
"learning_rate": 1.9532888007985408e-05,
|
267 |
+
"loss": 1.0957,
|
268 |
"step": 180
|
269 |
},
|
270 |
{
|
271 |
"epoch": 0.19351464435146443,
|
272 |
+
"grad_norm": 0.6000021573219635,
|
273 |
"learning_rate": 1.9476129425200935e-05,
|
274 |
+
"loss": 1.0938,
|
275 |
"step": 185
|
276 |
},
|
277 |
{
|
278 |
"epoch": 0.19874476987447698,
|
279 |
+
"grad_norm": 0.6537737768236229,
|
280 |
"learning_rate": 1.9416209571168648e-05,
|
281 |
"loss": 1.089,
|
282 |
"step": 190
|
283 |
},
|
284 |
{
|
285 |
"epoch": 0.20397489539748953,
|
286 |
+
"grad_norm": 0.5787562041762513,
|
287 |
"learning_rate": 1.9353148435369816e-05,
|
288 |
"loss": 1.0906,
|
289 |
"step": 195
|
290 |
},
|
291 |
{
|
292 |
"epoch": 0.20920502092050208,
|
293 |
+
"grad_norm": 0.5695490038983394,
|
294 |
"learning_rate": 1.9286967055228744e-05,
|
295 |
"loss": 1.093,
|
296 |
"step": 200
|
297 |
},
|
298 |
{
|
299 |
"epoch": 0.21443514644351463,
|
300 |
+
"grad_norm": 0.6745539583613525,
|
301 |
"learning_rate": 1.9217687509094596e-05,
|
302 |
"loss": 1.0759,
|
303 |
"step": 205
|
304 |
},
|
305 |
{
|
306 |
"epoch": 0.2196652719665272,
|
307 |
+
"grad_norm": 0.6099357395998225,
|
308 |
"learning_rate": 1.9145332908875984e-05,
|
309 |
"loss": 1.0911,
|
310 |
"step": 210
|
311 |
},
|
312 |
{
|
313 |
"epoch": 0.22489539748953974,
|
314 |
+
"grad_norm": 0.6131129695854803,
|
315 |
"learning_rate": 1.9069927392330742e-05,
|
316 |
"loss": 1.1017,
|
317 |
"step": 215
|
318 |
},
|
319 |
{
|
320 |
"epoch": 0.2301255230125523,
|
321 |
+
"grad_norm": 0.700006659493977,
|
322 |
"learning_rate": 1.89914961150135e-05,
|
323 |
"loss": 1.082,
|
324 |
"step": 220
|
325 |
},
|
326 |
{
|
327 |
"epoch": 0.23535564853556484,
|
328 |
+
"grad_norm": 0.532783820430617,
|
329 |
"learning_rate": 1.891006524188368e-05,
|
330 |
"loss": 1.0797,
|
331 |
"step": 225
|
332 |
},
|
333 |
{
|
334 |
"epoch": 0.2405857740585774,
|
335 |
+
"grad_norm": 0.6213200850147771,
|
336 |
"learning_rate": 1.8825661938576784e-05,
|
337 |
"loss": 1.085,
|
338 |
"step": 230
|
339 |
},
|
340 |
{
|
341 |
"epoch": 0.24581589958158995,
|
342 |
+
"grad_norm": 0.6011279894568407,
|
343 |
"learning_rate": 1.8738314362341828e-05,
|
344 |
+
"loss": 1.085,
|
345 |
"step": 235
|
346 |
},
|
347 |
{
|
348 |
"epoch": 0.2510460251046025,
|
349 |
+
"grad_norm": 0.5489112313533119,
|
350 |
"learning_rate": 1.864805165264799e-05,
|
351 |
+
"loss": 1.0952,
|
352 |
"step": 240
|
353 |
},
|
354 |
{
|
355 |
"epoch": 0.25627615062761505,
|
356 |
+
"grad_norm": 0.6812709510575856,
|
357 |
"learning_rate": 1.8554903921463578e-05,
|
358 |
"loss": 1.0889,
|
359 |
"step": 245
|
360 |
},
|
361 |
{
|
362 |
"epoch": 0.2615062761506276,
|
363 |
+
"grad_norm": 0.5886412319389147,
|
364 |
"learning_rate": 1.8458902243210558e-05,
|
365 |
"loss": 1.0822,
|
366 |
"step": 250
|
367 |
},
|
368 |
{
|
369 |
"epoch": 0.26673640167364016,
|
370 |
+
"grad_norm": 0.6816262740472222,
|
371 |
"learning_rate": 1.8360078644398028e-05,
|
372 |
"loss": 1.0788,
|
373 |
"step": 255
|
374 |
},
|
375 |
{
|
376 |
"epoch": 0.2719665271966527,
|
377 |
+
"grad_norm": 0.5947592675988007,
|
378 |
"learning_rate": 1.8258466092938042e-05,
|
379 |
"loss": 1.101,
|
380 |
"step": 260
|
381 |
},
|
382 |
{
|
383 |
"epoch": 0.27719665271966526,
|
384 |
+
"grad_norm": 0.57966086068774,
|
385 |
"learning_rate": 1.8154098487147405e-05,
|
386 |
"loss": 1.0776,
|
387 |
"step": 265
|
388 |
},
|
389 |
{
|
390 |
"epoch": 0.2824267782426778,
|
391 |
+
"grad_norm": 0.7809661648951529,
|
392 |
"learning_rate": 1.8047010644439074e-05,
|
393 |
"loss": 1.0838,
|
394 |
"step": 270
|
395 |
},
|
396 |
{
|
397 |
"epoch": 0.28765690376569036,
|
398 |
+
"grad_norm": 0.5662869436802427,
|
399 |
"learning_rate": 1.7937238289706927e-05,
|
400 |
+
"loss": 1.0822,
|
401 |
"step": 275
|
402 |
},
|
403 |
{
|
404 |
"epoch": 0.2928870292887029,
|
405 |
+
"grad_norm": 0.6222425392268467,
|
406 |
"learning_rate": 1.7824818043407828e-05,
|
407 |
+
"loss": 1.0641,
|
408 |
"step": 280
|
409 |
},
|
410 |
{
|
411 |
"epoch": 0.29811715481171547,
|
412 |
+
"grad_norm": 0.5849161394935216,
|
413 |
"learning_rate": 1.7709787409344893e-05,
|
414 |
+
"loss": 1.075,
|
415 |
"step": 285
|
416 |
},
|
417 |
{
|
418 |
"epoch": 0.303347280334728,
|
419 |
+
"grad_norm": 0.6269611916504507,
|
420 |
"learning_rate": 1.75921847621561e-05,
|
421 |
"loss": 1.0863,
|
422 |
"step": 290
|
423 |
},
|
424 |
{
|
425 |
"epoch": 0.30857740585774057,
|
426 |
+
"grad_norm": 0.6490170961985677,
|
427 |
"learning_rate": 1.7472049334512354e-05,
|
428 |
+
"loss": 1.0753,
|
429 |
"step": 295
|
430 |
},
|
431 |
{
|
432 |
"epoch": 0.3138075313807531,
|
433 |
+
"grad_norm": 0.5516576130404307,
|
434 |
"learning_rate": 1.7349421204029343e-05,
|
435 |
"loss": 1.0838,
|
436 |
"step": 300
|
437 |
},
|
438 |
{
|
439 |
"epoch": 0.3190376569037657,
|
440 |
+
"grad_norm": 0.590710588774648,
|
441 |
"learning_rate": 1.7224341279897473e-05,
|
442 |
"loss": 1.0937,
|
443 |
"step": 305
|
444 |
},
|
445 |
{
|
446 |
"epoch": 0.32426778242677823,
|
447 |
+
"grad_norm": 0.6160629723045913,
|
448 |
"learning_rate": 1.7096851289234448e-05,
|
449 |
"loss": 1.0861,
|
450 |
"step": 310
|
451 |
},
|
452 |
{
|
453 |
"epoch": 0.3294979079497908,
|
454 |
+
"grad_norm": 0.6756970289943374,
|
455 |
"learning_rate": 1.6966993763164928e-05,
|
456 |
"loss": 1.0863,
|
457 |
"step": 315
|
458 |
},
|
459 |
{
|
460 |
"epoch": 0.33472803347280333,
|
461 |
+
"grad_norm": 0.5500515778659105,
|
462 |
"learning_rate": 1.6834812022632e-05,
|
463 |
+
"loss": 1.0916,
|
464 |
"step": 320
|
465 |
},
|
466 |
{
|
467 |
"epoch": 0.3399581589958159,
|
468 |
+
"grad_norm": 0.5597424586372212,
|
469 |
"learning_rate": 1.670035016394513e-05,
|
470 |
"loss": 1.0875,
|
471 |
"step": 325
|
472 |
},
|
473 |
{
|
474 |
"epoch": 0.34518828451882844,
|
475 |
+
"grad_norm": 0.5731613149047409,
|
476 |
"learning_rate": 1.656365304406953e-05,
|
477 |
+
"loss": 1.0812,
|
478 |
"step": 330
|
479 |
},
|
480 |
{
|
481 |
"epoch": 0.350418410041841,
|
482 |
+
"grad_norm": 0.5799594878521428,
|
483 |
"learning_rate": 1.6424766265661648e-05,
|
484 |
"loss": 1.084,
|
485 |
"step": 335
|
486 |
},
|
487 |
{
|
488 |
"epoch": 0.35564853556485354,
|
489 |
+
"grad_norm": 0.6429338013882139,
|
490 |
"learning_rate": 1.6283736161855995e-05,
|
491 |
"loss": 1.0772,
|
492 |
"step": 340
|
493 |
},
|
494 |
{
|
495 |
"epoch": 0.3608786610878661,
|
496 |
+
"grad_norm": 0.5360765914109888,
|
497 |
"learning_rate": 1.614060978080823e-05,
|
498 |
+
"loss": 1.0928,
|
499 |
"step": 345
|
500 |
},
|
501 |
{
|
502 |
"epoch": 0.36610878661087864,
|
503 |
+
"grad_norm": 0.5564731824128625,
|
504 |
"learning_rate": 1.5995434869999723e-05,
|
505 |
"loss": 1.079,
|
506 |
"step": 350
|
507 |
},
|
508 |
{
|
509 |
"epoch": 0.3713389121338912,
|
510 |
+
"grad_norm": 0.561749976298582,
|
511 |
"learning_rate": 1.584825986030883e-05,
|
512 |
"loss": 1.0876,
|
513 |
"step": 355
|
514 |
},
|
515 |
{
|
516 |
"epoch": 0.37656903765690375,
|
517 |
+
"grad_norm": 0.5636262441838287,
|
518 |
"learning_rate": 1.5699133849854164e-05,
|
519 |
+
"loss": 1.0889,
|
520 |
"step": 360
|
521 |
},
|
522 |
{
|
523 |
"epoch": 0.3817991631799163,
|
524 |
+
"grad_norm": 0.5703557438762126,
|
525 |
"learning_rate": 1.5548106587615295e-05,
|
526 |
+
"loss": 1.067,
|
527 |
"step": 365
|
528 |
},
|
529 |
{
|
530 |
"epoch": 0.38702928870292885,
|
531 |
+
"grad_norm": 0.5519913100989867,
|
532 |
"learning_rate": 1.5395228456836298e-05,
|
533 |
"loss": 1.084,
|
534 |
"step": 370
|
535 |
},
|
536 |
{
|
537 |
"epoch": 0.3922594142259414,
|
538 |
+
"grad_norm": 0.6641842447650281,
|
539 |
"learning_rate": 1.524055045821773e-05,
|
540 |
"loss": 1.0915,
|
541 |
"step": 375
|
542 |
},
|
543 |
{
|
544 |
"epoch": 0.39748953974895396,
|
545 |
+
"grad_norm": 0.6318777460752142,
|
546 |
"learning_rate": 1.5084124192902612e-05,
|
547 |
"loss": 1.0788,
|
548 |
"step": 380
|
549 |
},
|
550 |
{
|
551 |
"epoch": 0.4027196652719665,
|
552 |
+
"grad_norm": 0.5871983626150223,
|
553 |
"learning_rate": 1.4926001845262081e-05,
|
554 |
"loss": 1.0831,
|
555 |
"step": 385
|
556 |
},
|
557 |
{
|
558 |
"epoch": 0.40794979079497906,
|
559 |
+
"grad_norm": 0.5413531651813962,
|
560 |
"learning_rate": 1.4766236165486526e-05,
|
561 |
"loss": 1.0798,
|
562 |
"step": 390
|
563 |
},
|
564 |
{
|
565 |
"epoch": 0.4131799163179916,
|
566 |
+
"grad_norm": 0.5573908575490272,
|
567 |
"learning_rate": 1.4604880451987895e-05,
|
568 |
"loss": 1.0763,
|
569 |
"step": 395
|
570 |
},
|
571 |
{
|
572 |
"epoch": 0.41841004184100417,
|
573 |
+
"grad_norm": 0.5194023684186305,
|
574 |
"learning_rate": 1.4441988533619182e-05,
|
575 |
"loss": 1.0735,
|
576 |
"step": 400
|
577 |
},
|
578 |
{
|
579 |
"epoch": 0.4236401673640167,
|
580 |
+
"grad_norm": 0.5830453281999414,
|
581 |
"learning_rate": 1.4277614751716911e-05,
|
582 |
"loss": 1.0821,
|
583 |
"step": 405
|
584 |
},
|
585 |
{
|
586 |
"epoch": 0.42887029288702927,
|
587 |
+
"grad_norm": 0.5637920375973177,
|
588 |
"learning_rate": 1.4111813941972672e-05,
|
589 |
"loss": 1.0961,
|
590 |
"step": 410
|
591 |
},
|
592 |
{
|
593 |
"epoch": 0.4341004184100418,
|
594 |
+
"grad_norm": 0.5841576073237601,
|
595 |
"learning_rate": 1.3944641416139737e-05,
|
596 |
+
"loss": 1.0817,
|
597 |
"step": 415
|
598 |
},
|
599 |
{
|
600 |
"epoch": 0.4393305439330544,
|
601 |
+
"grad_norm": 0.5585707054156458,
|
602 |
"learning_rate": 1.3776152943580846e-05,
|
603 |
+
"loss": 1.0913,
|
604 |
"step": 420
|
605 |
},
|
606 |
{
|
607 |
"epoch": 0.4445606694560669,
|
608 |
+
"grad_norm": 0.5469930648842268,
|
609 |
"learning_rate": 1.3606404732663365e-05,
|
610 |
+
"loss": 1.0698,
|
611 |
"step": 425
|
612 |
},
|
613 |
{
|
614 |
"epoch": 0.4497907949790795,
|
615 |
+
"grad_norm": 0.6030829852829179,
|
616 |
"learning_rate": 1.3435453412007949e-05,
|
617 |
+
"loss": 1.0679,
|
618 |
"step": 430
|
619 |
},
|
620 |
{
|
621 |
"epoch": 0.45502092050209203,
|
622 |
+
"grad_norm": 0.5093855577577417,
|
623 |
"learning_rate": 1.3263356011597045e-05,
|
624 |
+
"loss": 1.0757,
|
625 |
"step": 435
|
626 |
},
|
627 |
{
|
628 |
"epoch": 0.4602510460251046,
|
629 |
+
"grad_norm": 0.5491996878865052,
|
630 |
"learning_rate": 1.3090169943749475e-05,
|
631 |
+
"loss": 1.0787,
|
632 |
"step": 440
|
633 |
},
|
634 |
{
|
635 |
"epoch": 0.46548117154811713,
|
636 |
+
"grad_norm": 0.512248753718038,
|
637 |
"learning_rate": 1.2915952983967477e-05,
|
638 |
+
"loss": 1.0877,
|
639 |
"step": 445
|
640 |
},
|
641 |
{
|
642 |
"epoch": 0.4707112970711297,
|
643 |
+
"grad_norm": 0.5483058294847694,
|
644 |
"learning_rate": 1.2740763251662585e-05,
|
645 |
+
"loss": 1.1004,
|
646 |
"step": 450
|
647 |
},
|
648 |
{
|
649 |
"epoch": 0.47594142259414224,
|
650 |
+
"grad_norm": 0.6156670796243979,
|
651 |
"learning_rate": 1.2564659190766788e-05,
|
652 |
"loss": 1.0764,
|
653 |
"step": 455
|
654 |
},
|
655 |
{
|
656 |
"epoch": 0.4811715481171548,
|
657 |
+
"grad_norm": 0.5548599105126784,
|
658 |
"learning_rate": 1.2387699550235419e-05,
|
659 |
"loss": 1.0858,
|
660 |
"step": 460
|
661 |
},
|
662 |
{
|
663 |
"epoch": 0.48640167364016734,
|
664 |
+
"grad_norm": 0.4944121544407168,
|
665 |
"learning_rate": 1.2209943364448298e-05,
|
666 |
"loss": 1.0584,
|
667 |
"step": 465
|
668 |
},
|
669 |
{
|
670 |
"epoch": 0.4916317991631799,
|
671 |
+
"grad_norm": 0.552437443100853,
|
672 |
"learning_rate": 1.2031449933515625e-05,
|
673 |
"loss": 1.0687,
|
674 |
"step": 470
|
675 |
},
|
676 |
{
|
677 |
"epoch": 0.49686192468619245,
|
678 |
+
"grad_norm": 0.5439607969799337,
|
679 |
"learning_rate": 1.1852278803495267e-05,
|
680 |
+
"loss": 1.0574,
|
681 |
"step": 475
|
682 |
},
|
683 |
{
|
684 |
"epoch": 0.502092050209205,
|
685 |
+
"grad_norm": 0.644493491235614,
|
686 |
"learning_rate": 1.1672489746527979e-05,
|
687 |
"loss": 1.0758,
|
688 |
"step": 480
|
689 |
},
|
690 |
{
|
691 |
"epoch": 0.5073221757322176,
|
692 |
+
"grad_norm": 0.6395625397020802,
|
693 |
"learning_rate": 1.1492142740897188e-05,
|
694 |
"loss": 1.0677,
|
695 |
"step": 485
|
696 |
},
|
697 |
{
|
698 |
"epoch": 0.5125523012552301,
|
699 |
+
"grad_norm": 0.556432612437615,
|
700 |
"learning_rate": 1.1311297951020028e-05,
|
701 |
"loss": 1.0652,
|
702 |
"step": 490
|
703 |
},
|
704 |
{
|
705 |
"epoch": 0.5177824267782427,
|
706 |
+
"grad_norm": 0.7228902841727272,
|
707 |
"learning_rate": 1.1130015707376262e-05,
|
708 |
"loss": 1.0715,
|
709 |
"step": 495
|
710 |
},
|
711 |
{
|
712 |
"epoch": 0.5230125523012552,
|
713 |
+
"grad_norm": 0.5377283657779242,
|
714 |
"learning_rate": 1.0948356486381829e-05,
|
715 |
"loss": 1.0769,
|
716 |
"step": 500
|
717 |
},
|
718 |
{
|
719 |
"epoch": 0.5282426778242678,
|
720 |
+
"grad_norm": 0.5540329931082308,
|
721 |
"learning_rate": 1.0766380890213675e-05,
|
722 |
"loss": 1.0806,
|
723 |
"step": 505
|
724 |
},
|
725 |
{
|
726 |
"epoch": 0.5334728033472803,
|
727 |
+
"grad_norm": 0.5757757293198247,
|
728 |
"learning_rate": 1.0584149626592662e-05,
|
729 |
"loss": 1.056,
|
730 |
"step": 510
|
731 |
},
|
732 |
{
|
733 |
"epoch": 0.5387029288702929,
|
734 |
+
"grad_norm": 0.6168745200076227,
|
735 |
"learning_rate": 1.0401723488531238e-05,
|
736 |
+
"loss": 1.0683,
|
737 |
"step": 515
|
738 |
},
|
739 |
{
|
740 |
"epoch": 0.5439330543933054,
|
741 |
+
"grad_norm": 0.541484533483186,
|
742 |
"learning_rate": 1.0219163334052682e-05,
|
743 |
+
"loss": 1.0818,
|
744 |
"step": 520
|
745 |
},
|
746 |
{
|
747 |
"epoch": 0.549163179916318,
|
748 |
+
"grad_norm": 0.6087775414515405,
|
749 |
"learning_rate": 1.003653006588863e-05,
|
750 |
"loss": 1.0757,
|
751 |
"step": 525
|
752 |
},
|
753 |
{
|
754 |
"epoch": 0.5543933054393305,
|
755 |
+
"grad_norm": 0.514881446809141,
|
756 |
"learning_rate": 9.853884611161709e-06,
|
757 |
"loss": 1.047,
|
758 |
"step": 530
|
759 |
},
|
760 |
{
|
761 |
"epoch": 0.5596234309623431,
|
762 |
+
"grad_norm": 0.5141201235730006,
|
763 |
"learning_rate": 9.671287901060035e-06,
|
764 |
"loss": 1.064,
|
765 |
"step": 535
|
766 |
},
|
767 |
{
|
768 |
"epoch": 0.5648535564853556,
|
769 |
+
"grad_norm": 0.558678168889332,
|
770 |
"learning_rate": 9.48880085051033e-06,
|
771 |
"loss": 1.0734,
|
772 |
"step": 540
|
773 |
},
|
774 |
{
|
775 |
"epoch": 0.5700836820083682,
|
776 |
+
"grad_norm": 0.5515688630887745,
|
777 |
"learning_rate": 9.306484337856497e-06,
|
778 |
+
"loss": 1.0776,
|
779 |
"step": 545
|
780 |
},
|
781 |
{
|
782 |
"epoch": 0.5753138075313807,
|
783 |
+
"grad_norm": 0.5198540078292905,
|
784 |
"learning_rate": 9.124399184550377e-06,
|
785 |
+
"loss": 1.0818,
|
786 |
"step": 550
|
787 |
},
|
788 |
{
|
789 |
"epoch": 0.5805439330543933,
|
790 |
+
"grad_norm": 0.5376863184344793,
|
791 |
"learning_rate": 8.94260613486149e-06,
|
792 |
"loss": 1.0757,
|
793 |
"step": 555
|
794 |
},
|
795 |
{
|
796 |
"epoch": 0.5857740585774058,
|
797 |
+
"grad_norm": 0.5211594345995101,
|
798 |
"learning_rate": 8.76116583561252e-06,
|
799 |
+
"loss": 1.059,
|
800 |
"step": 560
|
801 |
},
|
802 |
{
|
803 |
"epoch": 0.5910041841004184,
|
804 |
+
"grad_norm": 0.5365035781726012,
|
805 |
"learning_rate": 8.580138815947336e-06,
|
806 |
"loss": 1.0651,
|
807 |
"step": 565
|
808 |
},
|
809 |
{
|
810 |
"epoch": 0.5962343096234309,
|
811 |
+
"grad_norm": 0.6305280632754058,
|
812 |
"learning_rate": 8.399585467138215e-06,
|
813 |
"loss": 1.0763,
|
814 |
"step": 570
|
815 |
},
|
816 |
{
|
817 |
"epoch": 0.6014644351464435,
|
818 |
+
"grad_norm": 0.6210565261791966,
|
819 |
"learning_rate": 8.219566022439128e-06,
|
820 |
"loss": 1.0792,
|
821 |
"step": 575
|
822 |
},
|
823 |
{
|
824 |
"epoch": 0.606694560669456,
|
825 |
+
"grad_norm": 0.5614735395071686,
|
826 |
"learning_rate": 8.040140536991688e-06,
|
827 |
+
"loss": 1.0768,
|
828 |
"step": 580
|
829 |
},
|
830 |
{
|
831 |
"epoch": 0.6119246861924686,
|
832 |
+
"grad_norm": 0.6033992631021817,
|
833 |
"learning_rate": 7.861368867790541e-06,
|
834 |
+
"loss": 1.0793,
|
835 |
"step": 585
|
836 |
},
|
837 |
{
|
838 |
"epoch": 0.6171548117154811,
|
839 |
+
"grad_norm": 0.5261950818886735,
|
840 |
"learning_rate": 7.683310653714857e-06,
|
841 |
"loss": 1.0723,
|
842 |
"step": 590
|
843 |
},
|
844 |
{
|
845 |
"epoch": 0.6223849372384938,
|
846 |
+
"grad_norm": 0.6939327155318498,
|
847 |
"learning_rate": 7.506025295632577e-06,
|
848 |
+
"loss": 1.0647,
|
849 |
"step": 595
|
850 |
},
|
851 |
{
|
852 |
"epoch": 0.6276150627615062,
|
853 |
+
"grad_norm": 0.5456155215789399,
|
854 |
"learning_rate": 7.329571936584072e-06,
|
855 |
"loss": 1.0566,
|
856 |
"step": 600
|
857 |
},
|
858 |
{
|
859 |
"epoch": 0.6328451882845189,
|
860 |
+
"grad_norm": 0.5593291123422247,
|
861 |
"learning_rate": 7.154009442051813e-06,
|
862 |
"loss": 1.0562,
|
863 |
"step": 605
|
864 |
},
|
865 |
{
|
866 |
"epoch": 0.6380753138075314,
|
867 |
+
"grad_norm": 0.5765286225968658,
|
868 |
"learning_rate": 6.979396380322621e-06,
|
869 |
"loss": 1.0744,
|
870 |
"step": 610
|
871 |
},
|
872 |
{
|
873 |
"epoch": 0.643305439330544,
|
874 |
+
"grad_norm": 0.5457349758684551,
|
875 |
"learning_rate": 6.805791002949085e-06,
|
876 |
"loss": 1.0688,
|
877 |
"step": 615
|
878 |
},
|
879 |
{
|
880 |
"epoch": 0.6485355648535565,
|
881 |
+
"grad_norm": 0.5037932117786891,
|
882 |
"learning_rate": 6.63325122531663e-06,
|
883 |
"loss": 1.0541,
|
884 |
"step": 620
|
885 |
},
|
886 |
{
|
887 |
"epoch": 0.6537656903765691,
|
888 |
+
"grad_norm": 0.5198402167489076,
|
889 |
"learning_rate": 6.461834607322719e-06,
|
890 |
"loss": 1.0547,
|
891 |
"step": 625
|
892 |
},
|
893 |
{
|
894 |
"epoch": 0.6589958158995816,
|
895 |
+
"grad_norm": 0.532907357801191,
|
896 |
"learning_rate": 6.291598334174685e-06,
|
897 |
"loss": 1.0572,
|
898 |
"step": 630
|
899 |
},
|
900 |
{
|
901 |
"epoch": 0.6642259414225942,
|
902 |
+
"grad_norm": 0.49718822363891185,
|
903 |
"learning_rate": 6.122599197312525e-06,
|
904 |
"loss": 1.0858,
|
905 |
"step": 635
|
906 |
},
|
907 |
{
|
908 |
"epoch": 0.6694560669456067,
|
909 |
+
"grad_norm": 0.5022400515016125,
|
910 |
"learning_rate": 5.954893575463064e-06,
|
911 |
"loss": 1.0637,
|
912 |
"step": 640
|
913 |
},
|
914 |
{
|
915 |
"epoch": 0.6746861924686193,
|
916 |
+
"grad_norm": 0.5200472376048936,
|
917 |
"learning_rate": 5.788537415831825e-06,
|
918 |
"loss": 1.055,
|
919 |
"step": 645
|
920 |
},
|
921 |
{
|
922 |
"epoch": 0.6799163179916318,
|
923 |
+
"grad_norm": 0.5001640685326711,
|
924 |
"learning_rate": 5.623586215438813e-06,
|
925 |
"loss": 1.0641,
|
926 |
"step": 650
|
927 |
},
|
928 |
{
|
929 |
"epoch": 0.6851464435146444,
|
930 |
+
"grad_norm": 0.5054182621927257,
|
931 |
"learning_rate": 5.460095002604533e-06,
|
932 |
"loss": 1.0477,
|
933 |
"step": 655
|
934 |
},
|
935 |
{
|
936 |
"epoch": 0.6903765690376569,
|
937 |
+
"grad_norm": 0.540901612852172,
|
938 |
"learning_rate": 5.298118318592316e-06,
|
939 |
"loss": 1.0654,
|
940 |
"step": 660
|
941 |
},
|
942 |
{
|
943 |
"epoch": 0.6956066945606695,
|
944 |
+
"grad_norm": 0.5013385695399603,
|
945 |
"learning_rate": 5.137710199413182e-06,
|
946 |
+
"loss": 1.0686,
|
947 |
"step": 665
|
948 |
},
|
949 |
{
|
950 |
"epoch": 0.700836820083682,
|
951 |
+
"grad_norm": 0.5598314300382486,
|
952 |
"learning_rate": 4.978924157799208e-06,
|
953 |
+
"loss": 1.0601,
|
954 |
"step": 670
|
955 |
},
|
956 |
{
|
957 |
"epoch": 0.7060669456066946,
|
958 |
+
"grad_norm": 0.48540319643618013,
|
959 |
"learning_rate": 4.821813165351495e-06,
|
960 |
"loss": 1.0581,
|
961 |
"step": 675
|
962 |
},
|
963 |
{
|
964 |
"epoch": 0.7112970711297071,
|
965 |
+
"grad_norm": 0.5008016797878718,
|
966 |
"learning_rate": 4.666429634868651e-06,
|
967 |
"loss": 1.0851,
|
968 |
"step": 680
|
969 |
},
|
970 |
{
|
971 |
"epoch": 0.7165271966527197,
|
972 |
+
"grad_norm": 0.4863424416345438,
|
973 |
"learning_rate": 4.512825402861686e-06,
|
974 |
"loss": 1.0552,
|
975 |
"step": 685
|
976 |
},
|
977 |
{
|
978 |
"epoch": 0.7217573221757322,
|
979 |
+
"grad_norm": 0.5716252229591956,
|
980 |
"learning_rate": 4.361051712261173e-06,
|
981 |
"loss": 1.0596,
|
982 |
"step": 690
|
983 |
},
|
984 |
{
|
985 |
"epoch": 0.7269874476987448,
|
986 |
+
"grad_norm": 0.5290120655638039,
|
987 |
"learning_rate": 4.211159195322404e-06,
|
988 |
+
"loss": 1.05,
|
989 |
"step": 695
|
990 |
},
|
991 |
{
|
992 |
"epoch": 0.7322175732217573,
|
993 |
+
"grad_norm": 0.5201644477315278,
|
994 |
"learning_rate": 4.063197856734295e-06,
|
995 |
"loss": 1.0783,
|
996 |
"step": 700
|
997 |
},
|
998 |
{
|
999 |
"epoch": 0.7374476987447699,
|
1000 |
+
"grad_norm": 0.5029601729726882,
|
1001 |
"learning_rate": 3.917217056937649e-06,
|
1002 |
"loss": 1.0636,
|
1003 |
"step": 705
|
1004 |
},
|
1005 |
{
|
1006 |
"epoch": 0.7426778242677824,
|
1007 |
+
"grad_norm": 0.5165741001138463,
|
1008 |
"learning_rate": 3.773265495658309e-06,
|
1009 |
"loss": 1.0655,
|
1010 |
"step": 710
|
1011 |
},
|
1012 |
{
|
1013 |
"epoch": 0.747907949790795,
|
1014 |
+
"grad_norm": 0.4952826724245214,
|
1015 |
"learning_rate": 3.6313911956607827e-06,
|
1016 |
+
"loss": 1.0854,
|
1017 |
"step": 715
|
1018 |
},
|
1019 |
{
|
1020 |
"epoch": 0.7531380753138075,
|
1021 |
+
"grad_norm": 0.5146856714184609,
|
1022 |
"learning_rate": 3.491641486727645e-06,
|
1023 |
"loss": 1.0684,
|
1024 |
"step": 720
|
1025 |
},
|
1026 |
{
|
1027 |
"epoch": 0.7583682008368201,
|
1028 |
+
"grad_norm": 0.5340447988495044,
|
1029 |
"learning_rate": 3.354062989870166e-06,
|
1030 |
+
"loss": 1.0505,
|
1031 |
"step": 725
|
1032 |
},
|
1033 |
{
|
1034 |
"epoch": 0.7635983263598326,
|
1035 |
+
"grad_norm": 0.5104098596156186,
|
1036 |
"learning_rate": 3.2187016017753714e-06,
|
1037 |
"loss": 1.066,
|
1038 |
"step": 730
|
1039 |
},
|
1040 |
{
|
1041 |
"epoch": 0.7688284518828452,
|
1042 |
+
"grad_norm": 0.5171571509002556,
|
1043 |
"learning_rate": 3.085602479494735e-06,
|
1044 |
"loss": 1.0602,
|
1045 |
"step": 735
|
1046 |
},
|
1047 |
{
|
1048 |
"epoch": 0.7740585774058577,
|
1049 |
+
"grad_norm": 0.5015256263302731,
|
1050 |
"learning_rate": 2.954810025379633e-06,
|
1051 |
"loss": 1.071,
|
1052 |
"step": 740
|
1053 |
},
|
1054 |
{
|
1055 |
"epoch": 0.7792887029288703,
|
1056 |
+
"grad_norm": 0.5031988802471105,
|
1057 |
"learning_rate": 2.826367872268543e-06,
|
1058 |
"loss": 1.0517,
|
1059 |
"step": 745
|
1060 |
},
|
1061 |
{
|
1062 |
"epoch": 0.7845188284518828,
|
1063 |
+
"grad_norm": 0.5291480112168645,
|
1064 |
"learning_rate": 2.700318868930977e-06,
|
1065 |
"loss": 1.0683,
|
1066 |
"step": 750
|
1067 |
},
|
1068 |
{
|
1069 |
"epoch": 0.7897489539748954,
|
1070 |
+
"grad_norm": 0.5301814683256604,
|
1071 |
"learning_rate": 2.576705065772985e-06,
|
1072 |
"loss": 1.0543,
|
1073 |
"step": 755
|
1074 |
},
|
1075 |
{
|
1076 |
"epoch": 0.7949790794979079,
|
1077 |
+
"grad_norm": 0.4858932325034876,
|
1078 |
"learning_rate": 2.455567700808974e-06,
|
1079 |
"loss": 1.0437,
|
1080 |
"step": 760
|
1081 |
},
|
1082 |
{
|
1083 |
"epoch": 0.8002092050209205,
|
1084 |
+
"grad_norm": 0.5226986538780068,
|
1085 |
"learning_rate": 2.3369471859045558e-06,
|
1086 |
"loss": 1.0637,
|
1087 |
"step": 765
|
1088 |
},
|
1089 |
{
|
1090 |
"epoch": 0.805439330543933,
|
1091 |
+
"grad_norm": 0.5018373716025176,
|
1092 |
"learning_rate": 2.2208830932950175e-06,
|
1093 |
+
"loss": 1.0558,
|
1094 |
"step": 770
|
1095 |
},
|
1096 |
{
|
1097 |
"epoch": 0.8106694560669456,
|
1098 |
+
"grad_norm": 0.5012264173655667,
|
1099 |
"learning_rate": 2.107414142383857e-06,
|
1100 |
"loss": 1.0636,
|
1101 |
"step": 775
|
1102 |
},
|
1103 |
{
|
1104 |
"epoch": 0.8158995815899581,
|
1105 |
+
"grad_norm": 0.5329303605558987,
|
1106 |
"learning_rate": 1.996578186825876e-06,
|
1107 |
"loss": 1.0652,
|
1108 |
"step": 780
|
1109 |
},
|
1110 |
{
|
1111 |
"epoch": 0.8211297071129707,
|
1112 |
+
"grad_norm": 0.5063762746528606,
|
1113 |
"learning_rate": 1.8884122018990292e-06,
|
1114 |
+
"loss": 1.0574,
|
1115 |
"step": 785
|
1116 |
},
|
1117 |
{
|
1118 |
"epoch": 0.8263598326359832,
|
1119 |
+
"grad_norm": 0.507270043815261,
|
1120 |
"learning_rate": 1.7829522721693738e-06,
|
1121 |
"loss": 1.0476,
|
1122 |
"step": 790
|
1123 |
},
|
1124 |
{
|
1125 |
"epoch": 0.8315899581589958,
|
1126 |
+
"grad_norm": 0.524242600565246,
|
1127 |
"learning_rate": 1.680233579453101e-06,
|
1128 |
"loss": 1.0604,
|
1129 |
"step": 795
|
1130 |
},
|
1131 |
{
|
1132 |
"epoch": 0.8368200836820083,
|
1133 |
+
"grad_norm": 0.5006885648328792,
|
1134 |
"learning_rate": 1.5802903910797584e-06,
|
1135 |
"loss": 1.0623,
|
1136 |
"step": 800
|
1137 |
},
|
1138 |
{
|
1139 |
"epoch": 0.8420502092050209,
|
1140 |
+
"grad_norm": 0.4983602387356114,
|
1141 |
"learning_rate": 1.483156048460561e-06,
|
1142 |
"loss": 1.0702,
|
1143 |
"step": 805
|
1144 |
},
|
1145 |
{
|
1146 |
"epoch": 0.8472803347280334,
|
1147 |
+
"grad_norm": 0.5141728007092325,
|
1148 |
"learning_rate": 1.3888629559655497e-06,
|
1149 |
"loss": 1.065,
|
1150 |
"step": 810
|
1151 |
},
|
1152 |
{
|
1153 |
"epoch": 0.852510460251046,
|
1154 |
+
"grad_norm": 0.4977890298090738,
|
1155 |
"learning_rate": 1.2974425701133941e-06,
|
1156 |
"loss": 1.0519,
|
1157 |
"step": 815
|
1158 |
},
|
1159 |
{
|
1160 |
"epoch": 0.8577405857740585,
|
1161 |
+
"grad_norm": 0.5073113010910021,
|
1162 |
"learning_rate": 1.2089253890773789e-06,
|
1163 |
"loss": 1.0592,
|
1164 |
"step": 820
|
1165 |
},
|
1166 |
{
|
1167 |
"epoch": 0.8629707112970711,
|
1168 |
+
"grad_norm": 0.4919709083678304,
|
1169 |
"learning_rate": 1.1233409425110952e-06,
|
1170 |
"loss": 1.0562,
|
1171 |
"step": 825
|
1172 |
},
|
1173 |
{
|
1174 |
"epoch": 0.8682008368200836,
|
1175 |
+
"grad_norm": 0.4868416396718501,
|
1176 |
"learning_rate": 1.0407177816972558e-06,
|
1177 |
"loss": 1.0469,
|
1178 |
"step": 830
|
1179 |
},
|
1180 |
{
|
1181 |
"epoch": 0.8734309623430963,
|
1182 |
+
"grad_norm": 0.5022480244107099,
|
1183 |
"learning_rate": 9.610834700228654e-07,
|
1184 |
+
"loss": 1.0775,
|
1185 |
"step": 835
|
1186 |
},
|
1187 |
{
|
1188 |
"epoch": 0.8786610878661087,
|
1189 |
+
"grad_norm": 0.5053021620216933,
|
1190 |
"learning_rate": 8.844645737839874e-07,
|
1191 |
+
"loss": 1.0567,
|
1192 |
"step": 840
|
1193 |
},
|
1194 |
{
|
1195 |
"epoch": 0.8838912133891214,
|
1196 |
+
"grad_norm": 0.4754078550874839,
|
1197 |
"learning_rate": 8.108866533231252e-07,
|
1198 |
+
"loss": 1.0428,
|
1199 |
"step": 845
|
1200 |
},
|
1201 |
{
|
1202 |
"epoch": 0.8891213389121339,
|
1203 |
+
"grad_norm": 0.5132996856432627,
|
1204 |
"learning_rate": 7.403742545021986e-07,
|
1205 |
"loss": 1.0559,
|
1206 |
"step": 850
|
1207 |
},
|
1208 |
{
|
1209 |
"epoch": 0.8943514644351465,
|
1210 |
+
"grad_norm": 0.4962332804391809,
|
1211 |
"learning_rate": 6.729509005139634e-07,
|
1212 |
"loss": 1.0415,
|
1213 |
"step": 855
|
1214 |
},
|
1215 |
{
|
1216 |
"epoch": 0.899581589958159,
|
1217 |
+
"grad_norm": 0.5018234838855186,
|
1218 |
"learning_rate": 6.086390840345758e-07,
|
1219 |
+
"loss": 1.0499,
|
1220 |
"step": 860
|
1221 |
},
|
1222 |
{
|
1223 |
"epoch": 0.9048117154811716,
|
1224 |
+
"grad_norm": 0.49057781568767944,
|
1225 |
"learning_rate": 5.474602597199607e-07,
|
1226 |
"loss": 1.0704,
|
1227 |
"step": 865
|
1228 |
},
|
1229 |
{
|
1230 |
"epoch": 0.9100418410041841,
|
1231 |
+
"grad_norm": 0.4970194272797173,
|
1232 |
"learning_rate": 4.894348370484648e-07,
|
1233 |
"loss": 1.0691,
|
1234 |
"step": 870
|
1235 |
},
|
1236 |
{
|
1237 |
"epoch": 0.9152719665271967,
|
1238 |
+
"grad_norm": 0.5221499540872924,
|
1239 |
"learning_rate": 4.34582173512168e-07,
|
1240 |
+
"loss": 1.0676,
|
1241 |
"step": 875
|
1242 |
},
|
1243 |
{
|
1244 |
"epoch": 0.9205020920502092,
|
1245 |
+
"grad_norm": 0.48846308721575726,
|
1246 |
"learning_rate": 3.8292056815916965e-07,
|
1247 |
"loss": 1.059,
|
1248 |
"step": 880
|
1249 |
},
|
1250 |
{
|
1251 |
"epoch": 0.9257322175732218,
|
1252 |
+
"grad_norm": 0.5300257475921499,
|
1253 |
"learning_rate": 3.344672554889428e-07,
|
1254 |
+
"loss": 1.0692,
|
1255 |
"step": 885
|
1256 |
},
|
1257 |
{
|
1258 |
"epoch": 0.9309623430962343,
|
1259 |
+
"grad_norm": 0.4959679907039524,
|
1260 |
"learning_rate": 2.8923839970285473e-07,
|
1261 |
"loss": 1.0592,
|
1262 |
"step": 890
|
1263 |
},
|
1264 |
{
|
1265 |
"epoch": 0.9361924686192469,
|
1266 |
+
"grad_norm": 0.49081091342104494,
|
1267 |
"learning_rate": 2.472490893117274e-07,
|
1268 |
"loss": 1.0458,
|
1269 |
"step": 895
|
1270 |
},
|
1271 |
{
|
1272 |
"epoch": 0.9414225941422594,
|
1273 |
+
"grad_norm": 0.5007780005543897,
|
1274 |
"learning_rate": 2.0851333210225032e-07,
|
1275 |
"loss": 1.0608,
|
1276 |
"step": 900
|
1277 |
},
|
1278 |
{
|
1279 |
"epoch": 0.946652719665272,
|
1280 |
+
"grad_norm": 0.48329063115650944,
|
1281 |
"learning_rate": 1.730440504639408e-07,
|
1282 |
+
"loss": 1.0579,
|
1283 |
"step": 905
|
1284 |
},
|
1285 |
{
|
1286 |
"epoch": 0.9518828451882845,
|
1287 |
+
"grad_norm": 0.48895599668986894,
|
1288 |
"learning_rate": 1.408530770781813e-07,
|
1289 |
+
"loss": 1.0527,
|
1290 |
"step": 910
|
1291 |
},
|
1292 |
{
|
1293 |
"epoch": 0.9571129707112971,
|
1294 |
+
"grad_norm": 0.4977013333272583,
|
1295 |
"learning_rate": 1.1195115097079268e-07,
|
1296 |
+
"loss": 1.0685,
|
1297 |
"step": 915
|
1298 |
},
|
1299 |
{
|
1300 |
"epoch": 0.9623430962343096,
|
1301 |
+
"grad_norm": 0.48972695236839464,
|
1302 |
"learning_rate": 8.634791392946429e-08,
|
1303 |
+
"loss": 1.0677,
|
1304 |
"step": 920
|
1305 |
},
|
1306 |
{
|
1307 |
"epoch": 0.9675732217573222,
|
1308 |
+
"grad_norm": 0.4907567001368972,
|
1309 |
"learning_rate": 6.405190728721033e-08,
|
1310 |
"loss": 1.0455,
|
1311 |
"step": 925
|
1312 |
},
|
1313 |
{
|
1314 |
"epoch": 0.9728033472803347,
|
1315 |
+
"grad_norm": 0.4772333774679202,
|
1316 |
"learning_rate": 4.5070569072952485e-08,
|
1317 |
+
"loss": 1.061,
|
1318 |
"step": 930
|
1319 |
},
|
1320 |
{
|
1321 |
"epoch": 0.9780334728033473,
|
1322 |
+
"grad_norm": 0.49589302731265184,
|
1323 |
"learning_rate": 2.9410231530168087e-08,
|
1324 |
+
"loss": 1.0499,
|
1325 |
"step": 935
|
1326 |
},
|
1327 |
{
|
1328 |
"epoch": 0.9832635983263598,
|
1329 |
+
"grad_norm": 0.47194869812634765,
|
1330 |
"learning_rate": 1.7076119004429958e-08,
|
1331 |
"loss": 1.0763,
|
1332 |
"step": 940
|
1333 |
},
|
1334 |
{
|
1335 |
"epoch": 0.9884937238493724,
|
1336 |
+
"grad_norm": 0.4944519541688841,
|
1337 |
"learning_rate": 8.072346200544979e-09,
|
1338 |
"loss": 1.0672,
|
1339 |
"step": 945
|
1340 |
},
|
1341 |
{
|
1342 |
"epoch": 0.9937238493723849,
|
1343 |
+
"grad_norm": 0.4887018274547105,
|
1344 |
"learning_rate": 2.401916809872118e-09,
|
1345 |
+
"loss": 1.0628,
|
1346 |
"step": 950
|
1347 |
},
|
1348 |
{
|
1349 |
"epoch": 0.9989539748953975,
|
1350 |
+
"grad_norm": 0.47672135170607305,
|
1351 |
"learning_rate": 6.672250828620996e-11,
|
1352 |
"loss": 1.0627,
|
1353 |
"step": 955
|
1354 |
},
|
1355 |
{
|
1356 |
"epoch": 1.0,
|
1357 |
+
"eval_loss": 1.064562439918518,
|
1358 |
+
"eval_runtime": 286.7813,
|
1359 |
+
"eval_samples_per_second": 47.21,
|
1360 |
+
"eval_steps_per_second": 0.739,
|
1361 |
"step": 956
|
1362 |
},
|
1363 |
{
|
1364 |
"epoch": 1.0,
|
1365 |
"step": 956,
|
1366 |
+
"total_flos": 500925122740224.0,
|
1367 |
+
"train_loss": 1.0766765792000743,
|
1368 |
+
"train_runtime": 10814.9645,
|
1369 |
+
"train_samples_per_second": 11.312,
|
1370 |
+
"train_steps_per_second": 0.088
|
1371 |
}
|
1372 |
],
|
1373 |
"logging_steps": 5,
|
|
|
1375 |
"num_input_tokens_seen": 0,
|
1376 |
"num_train_epochs": 1,
|
1377 |
"save_steps": 100,
|
1378 |
+
"total_flos": 500925122740224.0,
|
1379 |
"train_batch_size": 8,
|
1380 |
"trial_name": null,
|
1381 |
"trial_params": null
|