Training in progress, step 120000
Browse files- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/pytorch_model.bin +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/rng_state_4.pth +1 -1
- last-checkpoint/rng_state_5.pth +1 -1
- last-checkpoint/rng_state_6.pth +1 -1
- last-checkpoint/rng_state_7.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +203 -3
- pytorch_model.bin +1 -1
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 50044689
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f5c1a1f468fc2d342a23f1ac0d5752569966a310ca13ca1f923b43583a411e84
|
3 |
size 50044689
|
last-checkpoint/pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 25761253
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65a92e8166f8fa9a54e98da57dbdf5c7d7bdbe9bb899a7638fe74e0582796f58
|
3 |
size 25761253
|
last-checkpoint/rng_state_0.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44b70180124cd582ce1fec2362a6c7b1500fac5e37241e6e64ee705ea6e177b8
|
3 |
size 14503
|
last-checkpoint/rng_state_1.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44b70180124cd582ce1fec2362a6c7b1500fac5e37241e6e64ee705ea6e177b8
|
3 |
size 14503
|
last-checkpoint/rng_state_2.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44b70180124cd582ce1fec2362a6c7b1500fac5e37241e6e64ee705ea6e177b8
|
3 |
size 14503
|
last-checkpoint/rng_state_3.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44b70180124cd582ce1fec2362a6c7b1500fac5e37241e6e64ee705ea6e177b8
|
3 |
size 14503
|
last-checkpoint/rng_state_4.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44b70180124cd582ce1fec2362a6c7b1500fac5e37241e6e64ee705ea6e177b8
|
3 |
size 14503
|
last-checkpoint/rng_state_5.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44b70180124cd582ce1fec2362a6c7b1500fac5e37241e6e64ee705ea6e177b8
|
3 |
size 14503
|
last-checkpoint/rng_state_6.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44b70180124cd582ce1fec2362a6c7b1500fac5e37241e6e64ee705ea6e177b8
|
3 |
size 14503
|
last-checkpoint/rng_state_7.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44b70180124cd582ce1fec2362a6c7b1500fac5e37241e6e64ee705ea6e177b8
|
3 |
size 14503
|
last-checkpoint/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c534c69c702472a462ee5c5c8e40f09fd7295b5cb464c1c3364c271724c3d05b
|
3 |
size 623
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -2206,11 +2206,211 @@
|
|
2206 |
"eval_samples_per_second": 1041.083,
|
2207 |
"eval_steps_per_second": 16.316,
|
2208 |
"step": 110000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2209 |
}
|
2210 |
],
|
2211 |
"max_steps": 250000,
|
2212 |
"num_train_epochs": 12,
|
2213 |
-
"total_flos": 1.
|
2214 |
"trial_name": null,
|
2215 |
"trial_params": null
|
2216 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 5.353319057815845,
|
5 |
+
"global_step": 120000,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
2206 |
"eval_samples_per_second": 1041.083,
|
2207 |
"eval_steps_per_second": 16.316,
|
2208 |
"step": 110000
|
2209 |
+
},
|
2210 |
+
{
|
2211 |
+
"epoch": 4.93,
|
2212 |
+
"learning_rate": 0.00038495756080926107,
|
2213 |
+
"loss": 0.3933,
|
2214 |
+
"step": 110500
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 4.95,
|
2218 |
+
"learning_rate": 0.0003830777659310416,
|
2219 |
+
"loss": 0.3932,
|
2220 |
+
"step": 111000
|
2221 |
+
},
|
2222 |
+
{
|
2223 |
+
"epoch": 4.95,
|
2224 |
+
"eval_loss": 0.36492955684661865,
|
2225 |
+
"eval_runtime": 2.2207,
|
2226 |
+
"eval_samples_per_second": 1034.354,
|
2227 |
+
"eval_steps_per_second": 16.211,
|
2228 |
+
"step": 111000
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 4.97,
|
2232 |
+
"learning_rate": 0.0003811945556784033,
|
2233 |
+
"loss": 0.3931,
|
2234 |
+
"step": 111500
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 5.0,
|
2238 |
+
"learning_rate": 0.00037930801242906366,
|
2239 |
+
"loss": 0.3925,
|
2240 |
+
"step": 112000
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 5.0,
|
2244 |
+
"eval_loss": 0.36507272720336914,
|
2245 |
+
"eval_runtime": 2.3005,
|
2246 |
+
"eval_samples_per_second": 998.488,
|
2247 |
+
"eval_steps_per_second": 15.649,
|
2248 |
+
"step": 112000
|
2249 |
+
},
|
2250 |
+
{
|
2251 |
+
"epoch": 5.02,
|
2252 |
+
"learning_rate": 0.00037741821870653576,
|
2253 |
+
"loss": 0.3924,
|
2254 |
+
"step": 112500
|
2255 |
+
},
|
2256 |
+
{
|
2257 |
+
"epoch": 5.04,
|
2258 |
+
"learning_rate": 0.00037552525717651905,
|
2259 |
+
"loss": 0.3919,
|
2260 |
+
"step": 113000
|
2261 |
+
},
|
2262 |
+
{
|
2263 |
+
"epoch": 5.04,
|
2264 |
+
"eval_loss": 0.3649584650993347,
|
2265 |
+
"eval_runtime": 2.1025,
|
2266 |
+
"eval_samples_per_second": 1092.533,
|
2267 |
+
"eval_steps_per_second": 17.123,
|
2268 |
+
"step": 113000
|
2269 |
+
},
|
2270 |
+
{
|
2271 |
+
"epoch": 5.06,
|
2272 |
+
"learning_rate": 0.0003736292106432834,
|
2273 |
+
"loss": 0.3918,
|
2274 |
+
"step": 113500
|
2275 |
+
},
|
2276 |
+
{
|
2277 |
+
"epoch": 5.09,
|
2278 |
+
"learning_rate": 0.00037173016204604647,
|
2279 |
+
"loss": 0.3915,
|
2280 |
+
"step": 114000
|
2281 |
+
},
|
2282 |
+
{
|
2283 |
+
"epoch": 5.09,
|
2284 |
+
"eval_loss": 0.36355069279670715,
|
2285 |
+
"eval_runtime": 2.1224,
|
2286 |
+
"eval_samples_per_second": 1082.26,
|
2287 |
+
"eval_steps_per_second": 16.962,
|
2288 |
+
"step": 114000
|
2289 |
+
},
|
2290 |
+
{
|
2291 |
+
"epoch": 5.11,
|
2292 |
+
"learning_rate": 0.0003698281944553456,
|
2293 |
+
"loss": 0.3915,
|
2294 |
+
"step": 114500
|
2295 |
+
},
|
2296 |
+
{
|
2297 |
+
"epoch": 5.13,
|
2298 |
+
"learning_rate": 0.0003679233910694053,
|
2299 |
+
"loss": 0.3911,
|
2300 |
+
"step": 115000
|
2301 |
+
},
|
2302 |
+
{
|
2303 |
+
"epoch": 5.13,
|
2304 |
+
"eval_loss": 0.36546093225479126,
|
2305 |
+
"eval_runtime": 2.202,
|
2306 |
+
"eval_samples_per_second": 1043.148,
|
2307 |
+
"eval_steps_per_second": 16.349,
|
2308 |
+
"step": 115000
|
2309 |
+
},
|
2310 |
+
{
|
2311 |
+
"epoch": 5.15,
|
2312 |
+
"learning_rate": 0.000366015835210496,
|
2313 |
+
"loss": 0.3911,
|
2314 |
+
"step": 115500
|
2315 |
+
},
|
2316 |
+
{
|
2317 |
+
"epoch": 5.17,
|
2318 |
+
"learning_rate": 0.0003641056103212908,
|
2319 |
+
"loss": 0.3905,
|
2320 |
+
"step": 116000
|
2321 |
+
},
|
2322 |
+
{
|
2323 |
+
"epoch": 5.17,
|
2324 |
+
"eval_loss": 0.36497965455055237,
|
2325 |
+
"eval_runtime": 2.1541,
|
2326 |
+
"eval_samples_per_second": 1066.328,
|
2327 |
+
"eval_steps_per_second": 16.712,
|
2328 |
+
"step": 116000
|
2329 |
+
},
|
2330 |
+
{
|
2331 |
+
"epoch": 5.2,
|
2332 |
+
"learning_rate": 0.00036219279996121446,
|
2333 |
+
"loss": 0.3904,
|
2334 |
+
"step": 116500
|
2335 |
+
},
|
2336 |
+
{
|
2337 |
+
"epoch": 5.22,
|
2338 |
+
"learning_rate": 0.0003602774878027888,
|
2339 |
+
"loss": 0.3905,
|
2340 |
+
"step": 117000
|
2341 |
+
},
|
2342 |
+
{
|
2343 |
+
"epoch": 5.22,
|
2344 |
+
"eval_loss": 0.40538668632507324,
|
2345 |
+
"eval_runtime": 2.1607,
|
2346 |
+
"eval_samples_per_second": 1063.067,
|
2347 |
+
"eval_steps_per_second": 16.661,
|
2348 |
+
"step": 117000
|
2349 |
+
},
|
2350 |
+
{
|
2351 |
+
"epoch": 5.24,
|
2352 |
+
"learning_rate": 0.00035835975762797245,
|
2353 |
+
"loss": 0.3917,
|
2354 |
+
"step": 117500
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 5.26,
|
2358 |
+
"learning_rate": 0.0003564396933244957,
|
2359 |
+
"loss": 0.3894,
|
2360 |
+
"step": 118000
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 5.26,
|
2364 |
+
"eval_loss": 0.36090952157974243,
|
2365 |
+
"eval_runtime": 2.2235,
|
2366 |
+
"eval_samples_per_second": 1033.034,
|
2367 |
+
"eval_steps_per_second": 16.19,
|
2368 |
+
"step": 118000
|
2369 |
+
},
|
2370 |
+
{
|
2371 |
+
"epoch": 5.29,
|
2372 |
+
"learning_rate": 0.0003545173788821915,
|
2373 |
+
"loss": 0.3893,
|
2374 |
+
"step": 118500
|
2375 |
+
},
|
2376 |
+
{
|
2377 |
+
"epoch": 5.31,
|
2378 |
+
"learning_rate": 0.00035259289838932104,
|
2379 |
+
"loss": 0.3889,
|
2380 |
+
"step": 119000
|
2381 |
+
},
|
2382 |
+
{
|
2383 |
+
"epoch": 5.31,
|
2384 |
+
"eval_loss": 0.3599041998386383,
|
2385 |
+
"eval_runtime": 2.1374,
|
2386 |
+
"eval_samples_per_second": 1074.669,
|
2387 |
+
"eval_steps_per_second": 16.843,
|
2388 |
+
"step": 119000
|
2389 |
+
},
|
2390 |
+
{
|
2391 |
+
"epoch": 5.33,
|
2392 |
+
"learning_rate": 0.0003506663360288954,
|
2393 |
+
"loss": 0.3889,
|
2394 |
+
"step": 119500
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 5.35,
|
2398 |
+
"learning_rate": 0.00034873777607499334,
|
2399 |
+
"loss": 0.3888,
|
2400 |
+
"step": 120000
|
2401 |
+
},
|
2402 |
+
{
|
2403 |
+
"epoch": 5.35,
|
2404 |
+
"eval_loss": 0.3593074679374695,
|
2405 |
+
"eval_runtime": 2.1457,
|
2406 |
+
"eval_samples_per_second": 1070.506,
|
2407 |
+
"eval_steps_per_second": 16.778,
|
2408 |
+
"step": 120000
|
2409 |
}
|
2410 |
],
|
2411 |
"max_steps": 250000,
|
2412 |
"num_train_epochs": 12,
|
2413 |
+
"total_flos": 1.9219803744787812e+21,
|
2414 |
"trial_name": null,
|
2415 |
"trial_params": null
|
2416 |
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 25761253
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65a92e8166f8fa9a54e98da57dbdf5c7d7bdbe9bb899a7638fe74e0582796f58
|
3 |
size 25761253
|