Commit
•
faadcb1
1
Parent(s):
2c0a422
Training in progress, step 22000
Browse files- adapter_model.bin +1 -1
- {checkpoint-18000 → checkpoint-21000/adapter_model}/README.md +0 -0
- {checkpoint-18000 → checkpoint-21000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-18000 → checkpoint-21000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-18000/adapter_model → checkpoint-22000}/README.md +0 -0
- {checkpoint-18000/adapter_model → checkpoint-22000}/adapter_config.json +0 -0
- {checkpoint-18000/adapter_model → checkpoint-22000}/adapter_model.bin +1 -1
- {checkpoint-18000 → checkpoint-22000}/optimizer.pt +1 -1
- {checkpoint-18000 → checkpoint-22000}/rng_state.pth +1 -1
- {checkpoint-18000 → checkpoint-22000}/scheduler.pt +1 -1
- {checkpoint-18000 → checkpoint-22000}/trainer_state.json +277 -5
- {checkpoint-18000 → checkpoint-22000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d90531691a4bd22a7fc951651aaad0da90ed5d3154cec0eed731f4f8c1da5c1
|
| 3 |
size 500897101
|
{checkpoint-18000 → checkpoint-21000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-18000 → checkpoint-21000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-18000 → checkpoint-21000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f715fb1f731563489cd7a96170934d1a2704e0ffdab19dbb6afc7d46ea57e62
|
| 3 |
size 500897101
|
{checkpoint-18000/adapter_model → checkpoint-22000}/README.md
RENAMED
|
File without changes
|
{checkpoint-18000/adapter_model → checkpoint-22000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-18000/adapter_model → checkpoint-22000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d90531691a4bd22a7fc951651aaad0da90ed5d3154cec0eed731f4f8c1da5c1
|
| 3 |
size 500897101
|
{checkpoint-18000 → checkpoint-22000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32d1d8e706f824a607393e4770716b98e3e14380a36b3d7dd013d1899e28d004
|
| 3 |
size 1001723453
|
{checkpoint-18000 → checkpoint-22000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b542bc6d6985311dc1f25608b1f96d66e1395f22a48f78d5c5c683d2d8fdde5b
|
| 3 |
size 14575
|
{checkpoint-18000 → checkpoint-22000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1276f4059dba941483dcef8d87faad5b272a9ed0c06d8c247d04b1898961557
|
| 3 |
size 627
|
{checkpoint-18000 → checkpoint-22000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 0.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -1230,11 +1230,283 @@
|
|
| 1230 |
"eval_samples_per_second": 0.416,
|
| 1231 |
"eval_steps_per_second": 0.416,
|
| 1232 |
"step": 18000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1233 |
}
|
| 1234 |
],
|
| 1235 |
"max_steps": 80463,
|
| 1236 |
"num_train_epochs": 3,
|
| 1237 |
-
"total_flos":
|
| 1238 |
"trial_name": null,
|
| 1239 |
"trial_params": null
|
| 1240 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.6218891143798828,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-22000",
|
| 4 |
+
"epoch": 0.8202527869952649,
|
| 5 |
+
"global_step": 22000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1230 |
"eval_samples_per_second": 0.416,
|
| 1231 |
"eval_steps_per_second": 0.416,
|
| 1232 |
"step": 18000
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 0.67,
|
| 1236 |
+
"learning_rate": 0.0001760708303094572,
|
| 1237 |
+
"loss": 0.6183,
|
| 1238 |
+
"step": 18100
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.68,
|
| 1242 |
+
"learning_rate": 0.00017581678764889324,
|
| 1243 |
+
"loss": 0.6116,
|
| 1244 |
+
"step": 18200
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 0.68,
|
| 1248 |
+
"learning_rate": 0.00017556158892909567,
|
| 1249 |
+
"loss": 0.6406,
|
| 1250 |
+
"step": 18300
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.69,
|
| 1254 |
+
"learning_rate": 0.00017530523804135085,
|
| 1255 |
+
"loss": 0.6223,
|
| 1256 |
+
"step": 18400
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.69,
|
| 1260 |
+
"learning_rate": 0.00017504773889451361,
|
| 1261 |
+
"loss": 0.628,
|
| 1262 |
+
"step": 18500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.69,
|
| 1266 |
+
"learning_rate": 0.00017478909541494736,
|
| 1267 |
+
"loss": 0.6173,
|
| 1268 |
+
"step": 18600
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.7,
|
| 1272 |
+
"learning_rate": 0.00017452931154646444,
|
| 1273 |
+
"loss": 0.61,
|
| 1274 |
+
"step": 18700
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.7,
|
| 1278 |
+
"learning_rate": 0.00017426839125026598,
|
| 1279 |
+
"loss": 0.5959,
|
| 1280 |
+
"step": 18800
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.7,
|
| 1284 |
+
"learning_rate": 0.00017400633850488128,
|
| 1285 |
+
"loss": 0.5979,
|
| 1286 |
+
"step": 18900
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 0.71,
|
| 1290 |
+
"learning_rate": 0.00017374315730610745,
|
| 1291 |
+
"loss": 0.6161,
|
| 1292 |
+
"step": 19000
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.71,
|
| 1296 |
+
"eval_loss": 0.6378119587898254,
|
| 1297 |
+
"eval_runtime": 1283.5987,
|
| 1298 |
+
"eval_samples_per_second": 0.422,
|
| 1299 |
+
"eval_steps_per_second": 0.422,
|
| 1300 |
+
"step": 19000
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 0.71,
|
| 1304 |
+
"learning_rate": 0.00017347885166694825,
|
| 1305 |
+
"loss": 0.6213,
|
| 1306 |
+
"step": 19100
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.72,
|
| 1310 |
+
"learning_rate": 0.00017321342561755297,
|
| 1311 |
+
"loss": 0.6217,
|
| 1312 |
+
"step": 19200
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.72,
|
| 1316 |
+
"learning_rate": 0.00017294688320515506,
|
| 1317 |
+
"loss": 0.6127,
|
| 1318 |
+
"step": 19300
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.72,
|
| 1322 |
+
"learning_rate": 0.00017267922849401024,
|
| 1323 |
+
"loss": 0.6145,
|
| 1324 |
+
"step": 19400
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 0.73,
|
| 1328 |
+
"learning_rate": 0.00017241046556533472,
|
| 1329 |
+
"loss": 0.5936,
|
| 1330 |
+
"step": 19500
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.73,
|
| 1334 |
+
"learning_rate": 0.0001721405985172428,
|
| 1335 |
+
"loss": 0.6273,
|
| 1336 |
+
"step": 19600
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.73,
|
| 1340 |
+
"learning_rate": 0.0001718696314646846,
|
| 1341 |
+
"loss": 0.6059,
|
| 1342 |
+
"step": 19700
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 0.74,
|
| 1346 |
+
"learning_rate": 0.000171597568539383,
|
| 1347 |
+
"loss": 0.5934,
|
| 1348 |
+
"step": 19800
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.74,
|
| 1352 |
+
"learning_rate": 0.000171324413889771,
|
| 1353 |
+
"loss": 0.6243,
|
| 1354 |
+
"step": 19900
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.75,
|
| 1358 |
+
"learning_rate": 0.00017105017168092808,
|
| 1359 |
+
"loss": 0.6164,
|
| 1360 |
+
"step": 20000
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.75,
|
| 1364 |
+
"eval_loss": 0.6324757933616638,
|
| 1365 |
+
"eval_runtime": 1266.6769,
|
| 1366 |
+
"eval_samples_per_second": 0.428,
|
| 1367 |
+
"eval_steps_per_second": 0.428,
|
| 1368 |
+
"step": 20000
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.75,
|
| 1372 |
+
"learning_rate": 0.0001707748460945171,
|
| 1373 |
+
"loss": 0.5953,
|
| 1374 |
+
"step": 20100
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.75,
|
| 1378 |
+
"learning_rate": 0.0001704984413287202,
|
| 1379 |
+
"loss": 0.6329,
|
| 1380 |
+
"step": 20200
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.76,
|
| 1384 |
+
"learning_rate": 0.00017022096159817493,
|
| 1385 |
+
"loss": 0.6227,
|
| 1386 |
+
"step": 20300
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 0.76,
|
| 1390 |
+
"learning_rate": 0.00016994241113391003,
|
| 1391 |
+
"loss": 0.6022,
|
| 1392 |
+
"step": 20400
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 0.76,
|
| 1396 |
+
"learning_rate": 0.0001696627941832808,
|
| 1397 |
+
"loss": 0.604,
|
| 1398 |
+
"step": 20500
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 0.77,
|
| 1402 |
+
"learning_rate": 0.0001693821150099044,
|
| 1403 |
+
"loss": 0.6101,
|
| 1404 |
+
"step": 20600
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.77,
|
| 1408 |
+
"learning_rate": 0.00016910037789359485,
|
| 1409 |
+
"loss": 0.6242,
|
| 1410 |
+
"step": 20700
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.78,
|
| 1414 |
+
"learning_rate": 0.00016881758713029776,
|
| 1415 |
+
"loss": 0.6096,
|
| 1416 |
+
"step": 20800
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.78,
|
| 1420 |
+
"learning_rate": 0.0001685337470320248,
|
| 1421 |
+
"loss": 0.5948,
|
| 1422 |
+
"step": 20900
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 0.78,
|
| 1426 |
+
"learning_rate": 0.0001682488619267879,
|
| 1427 |
+
"loss": 0.5911,
|
| 1428 |
+
"step": 21000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 0.78,
|
| 1432 |
+
"eval_loss": 0.6282580494880676,
|
| 1433 |
+
"eval_runtime": 1313.1215,
|
| 1434 |
+
"eval_samples_per_second": 0.413,
|
| 1435 |
+
"eval_steps_per_second": 0.413,
|
| 1436 |
+
"step": 21000
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 0.79,
|
| 1440 |
+
"learning_rate": 0.0001679629361585335,
|
| 1441 |
+
"loss": 0.5716,
|
| 1442 |
+
"step": 21100
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.79,
|
| 1446 |
+
"learning_rate": 0.00016767597408707594,
|
| 1447 |
+
"loss": 0.5957,
|
| 1448 |
+
"step": 21200
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 0.79,
|
| 1452 |
+
"learning_rate": 0.00016738798008803128,
|
| 1453 |
+
"loss": 0.6308,
|
| 1454 |
+
"step": 21300
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 0.8,
|
| 1458 |
+
"learning_rate": 0.00016709895855275048,
|
| 1459 |
+
"loss": 0.5891,
|
| 1460 |
+
"step": 21400
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.8,
|
| 1464 |
+
"learning_rate": 0.00016680891388825243,
|
| 1465 |
+
"loss": 0.6104,
|
| 1466 |
+
"step": 21500
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.81,
|
| 1470 |
+
"learning_rate": 0.00016651785051715674,
|
| 1471 |
+
"loss": 0.6344,
|
| 1472 |
+
"step": 21600
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.81,
|
| 1476 |
+
"learning_rate": 0.0001662257728776163,
|
| 1477 |
+
"loss": 0.604,
|
| 1478 |
+
"step": 21700
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 0.81,
|
| 1482 |
+
"learning_rate": 0.0001659326854232497,
|
| 1483 |
+
"loss": 0.6066,
|
| 1484 |
+
"step": 21800
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 0.82,
|
| 1488 |
+
"learning_rate": 0.0001656385926230732,
|
| 1489 |
+
"loss": 0.6324,
|
| 1490 |
+
"step": 21900
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.82,
|
| 1494 |
+
"learning_rate": 0.00016534349896143264,
|
| 1495 |
+
"loss": 0.5819,
|
| 1496 |
+
"step": 22000
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 0.82,
|
| 1500 |
+
"eval_loss": 0.6218891143798828,
|
| 1501 |
+
"eval_runtime": 1296.6038,
|
| 1502 |
+
"eval_samples_per_second": 0.418,
|
| 1503 |
+
"eval_steps_per_second": 0.418,
|
| 1504 |
+
"step": 22000
|
| 1505 |
}
|
| 1506 |
],
|
| 1507 |
"max_steps": 80463,
|
| 1508 |
"num_train_epochs": 3,
|
| 1509 |
+
"total_flos": 6.16734700814082e+18,
|
| 1510 |
"trial_name": null,
|
| 1511 |
"trial_params": null
|
| 1512 |
}
|
{checkpoint-18000 → checkpoint-22000}/training_args.bin
RENAMED
|
File without changes
|