Gayuru commited on
Commit
8be4a79
·
verified ·
1 Parent(s): 4013375

Training in progress, step 6000, checkpoint

Browse files
last-checkpoint/adapter_config.json CHANGED
@@ -34,11 +34,11 @@
34
  "rank_pattern": {},
35
  "revision": null,
36
  "target_modules": [
 
37
  "k_proj",
 
38
  "fc1",
39
- "fc2",
40
  "v_proj",
41
- "q_proj",
42
  "out_proj"
43
  ],
44
  "target_parameters": null,
 
34
  "rank_pattern": {},
35
  "revision": null,
36
  "target_modules": [
37
+ "fc2",
38
  "k_proj",
39
+ "q_proj",
40
  "fc1",
 
41
  "v_proj",
 
42
  "out_proj"
43
  ],
44
  "target_parameters": null,
last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:92cd05ecee776906acb4650b77e0c0e14681497c2a69ba72f53b7851a461c12e
3
  size 138520208
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0fb5500ae69f84a76754308d009df28141989c068177dfc386cceaf38f706cd
3
  size 138520208
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:822e27e6d06957f87b735c3e4e326c3b4c561dda34862e84252f510fe1ac8df1
3
- size 71193893
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d92f6982d4701c82d6cadd609c04f6a2ea079df652dece21ea84925951899849
3
+ size 73194211
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9d4be0365aabea1146efe489d2cc1a1afa5409c644730f2143a7027a00d20fa6
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:835efd0bf454549ca0472460674f59376fff3bc60ee76965c29d197e60989b9f
3
  size 14645
last-checkpoint/scaler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4c3f6b7af54b016b94bdaea40f9f685f706689b3824f25ae815d516be3f1461
3
  size 1383
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:540963318ae696eaccffdcc03357c38c2cfd8ac95fc4b8585222eadae778dddb
3
  size 1383
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:337f044b6ecc4905ff989d8975d0131db7133fae1d4f5b0d8b8e8178ed6340ea
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6773923f1419f113ca4d0492a6dd5c6af82b2b9307b3f6140af468e27a8fef36
3
  size 1465
last-checkpoint/trainer_state.json CHANGED
@@ -2,9 +2,9 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.837331206199077,
6
  "eval_steps": 1000,
7
- "global_step": 3674,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -1078,6 +1078,712 @@
1078
  "learning_rate": 0.00014679111385374884,
1079
  "loss": 0.1867,
1080
  "step": 3650
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1081
  }
1082
  ],
1083
  "logging_steps": 25,
@@ -1092,12 +1798,12 @@
1092
  "should_evaluate": false,
1093
  "should_log": false,
1094
  "should_save": true,
1095
- "should_training_stop": true
1096
  },
1097
  "attributes": {}
1098
  }
1099
  },
1100
- "total_flos": 6.292434752372736e+19,
1101
  "train_batch_size": 4,
1102
  "trial_name": null,
1103
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 1.3676143809469545,
6
  "eval_steps": 1000,
7
+ "global_step": 6000,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
1078
  "learning_rate": 0.00014679111385374884,
1079
  "loss": 0.1867,
1080
  "step": 3650
1081
+ },
1082
+ {
1083
+ "epoch": 0.8375591134408296,
1084
+ "grad_norm": NaN,
1085
+ "learning_rate": 0.00014640543042270904,
1086
+ "loss": 0.5879,
1087
+ "step": 3675
1088
+ },
1089
+ {
1090
+ "epoch": 0.8432567944846447,
1091
+ "grad_norm": 2.241427421569824,
1092
+ "learning_rate": 0.00014605060166615242,
1093
+ "loss": 0.6686,
1094
+ "step": 3700
1095
+ },
1096
+ {
1097
+ "epoch": 0.8489544755284599,
1098
+ "grad_norm": 4.021817207336426,
1099
+ "learning_rate": 0.00014566491823511264,
1100
+ "loss": 1.2459,
1101
+ "step": 3725
1102
+ },
1103
+ {
1104
+ "epoch": 0.8546521565722751,
1105
+ "grad_norm": 4.461009979248047,
1106
+ "learning_rate": 0.0001452792348040728,
1107
+ "loss": 1.0951,
1108
+ "step": 3750
1109
+ },
1110
+ {
1111
+ "epoch": 0.8603498376160903,
1112
+ "grad_norm": 5.371738910675049,
1113
+ "learning_rate": 0.00014489355137303303,
1114
+ "loss": 1.0724,
1115
+ "step": 3775
1116
+ },
1117
+ {
1118
+ "epoch": 0.8660475186599054,
1119
+ "grad_norm": 3.075195074081421,
1120
+ "learning_rate": 0.00014450786794199322,
1121
+ "loss": 1.1333,
1122
+ "step": 3800
1123
+ },
1124
+ {
1125
+ "epoch": 0.8717451997037206,
1126
+ "grad_norm": 4.584661483764648,
1127
+ "learning_rate": 0.00014412218451095342,
1128
+ "loss": 0.9828,
1129
+ "step": 3825
1130
+ },
1131
+ {
1132
+ "epoch": 0.8774428807475357,
1133
+ "grad_norm": 4.700824737548828,
1134
+ "learning_rate": 0.0001437365010799136,
1135
+ "loss": 1.0094,
1136
+ "step": 3850
1137
+ },
1138
+ {
1139
+ "epoch": 0.8831405617913509,
1140
+ "grad_norm": 3.36844539642334,
1141
+ "learning_rate": 0.0001433508176488738,
1142
+ "loss": 1.0466,
1143
+ "step": 3875
1144
+ },
1145
+ {
1146
+ "epoch": 0.8888382428351661,
1147
+ "grad_norm": 3.3631935119628906,
1148
+ "learning_rate": 0.00014296513421783403,
1149
+ "loss": 0.9994,
1150
+ "step": 3900
1151
+ },
1152
+ {
1153
+ "epoch": 0.8945359238789813,
1154
+ "grad_norm": 6.935356616973877,
1155
+ "learning_rate": 0.0001425794507867942,
1156
+ "loss": 1.0332,
1157
+ "step": 3925
1158
+ },
1159
+ {
1160
+ "epoch": 0.9002336049227965,
1161
+ "grad_norm": 3.6604976654052734,
1162
+ "learning_rate": 0.0001421937673557544,
1163
+ "loss": 0.9961,
1164
+ "step": 3950
1165
+ },
1166
+ {
1167
+ "epoch": 0.9059312859666115,
1168
+ "grad_norm": 4.015224933624268,
1169
+ "learning_rate": 0.0001418080839247146,
1170
+ "loss": 0.9397,
1171
+ "step": 3975
1172
+ },
1173
+ {
1174
+ "epoch": 0.9116289670104267,
1175
+ "grad_norm": 6.816766262054443,
1176
+ "learning_rate": 0.00014142240049367478,
1177
+ "loss": 1.0108,
1178
+ "step": 4000
1179
+ },
1180
+ {
1181
+ "epoch": 0.9116289670104267,
1182
+ "eval_cer": 7.395053725424049,
1183
+ "eval_cer_en": 3.032200007825032,
1184
+ "eval_cer_si": 10.433286407007666,
1185
+ "eval_cer_ta": 8.719674761439446,
1186
+ "eval_loss": 0.2628401815891266,
1187
+ "eval_runtime": 115.4592,
1188
+ "eval_samples_per_second": 6.496,
1189
+ "eval_steps_per_second": 0.814,
1190
+ "eval_wer": 27.847510316482545,
1191
+ "eval_wer_en": 5.0041356492969395,
1192
+ "eval_wer_si": 47.310847766636286,
1193
+ "eval_wer_ta": 31.227547533514414,
1194
+ "step": 4000
1195
+ },
1196
+ {
1197
+ "epoch": 0.9173266480542419,
1198
+ "grad_norm": 3.8277945518493652,
1199
+ "learning_rate": 0.000141036717062635,
1200
+ "loss": 1.0051,
1201
+ "step": 4025
1202
+ },
1203
+ {
1204
+ "epoch": 0.9230243290980571,
1205
+ "grad_norm": 3.6770007610321045,
1206
+ "learning_rate": 0.0001406510336315952,
1207
+ "loss": 1.0117,
1208
+ "step": 4050
1209
+ },
1210
+ {
1211
+ "epoch": 0.9287220101418723,
1212
+ "grad_norm": 5.993704795837402,
1213
+ "learning_rate": 0.0001402653502005554,
1214
+ "loss": 0.9527,
1215
+ "step": 4075
1216
+ },
1217
+ {
1218
+ "epoch": 0.9344196911856875,
1219
+ "grad_norm": 3.0133066177368164,
1220
+ "learning_rate": 0.00013987966676951558,
1221
+ "loss": 0.9444,
1222
+ "step": 4100
1223
+ },
1224
+ {
1225
+ "epoch": 0.9401173722295026,
1226
+ "grad_norm": 5.697484970092773,
1227
+ "learning_rate": 0.00013949398333847578,
1228
+ "loss": 1.0819,
1229
+ "step": 4125
1230
+ },
1231
+ {
1232
+ "epoch": 0.9458150532733177,
1233
+ "grad_norm": 2.8382954597473145,
1234
+ "learning_rate": 0.000139108299907436,
1235
+ "loss": 0.9951,
1236
+ "step": 4150
1237
+ },
1238
+ {
1239
+ "epoch": 0.9515127343171329,
1240
+ "grad_norm": 5.698050022125244,
1241
+ "learning_rate": 0.00013872261647639617,
1242
+ "loss": 0.9218,
1243
+ "step": 4175
1244
+ },
1245
+ {
1246
+ "epoch": 0.9572104153609481,
1247
+ "grad_norm": 5.912572860717773,
1248
+ "learning_rate": 0.0001383369330453564,
1249
+ "loss": 0.979,
1250
+ "step": 4200
1251
+ },
1252
+ {
1253
+ "epoch": 0.9629080964047633,
1254
+ "grad_norm": 5.067142486572266,
1255
+ "learning_rate": 0.00013795124961431658,
1256
+ "loss": 0.9073,
1257
+ "step": 4225
1258
+ },
1259
+ {
1260
+ "epoch": 0.9686057774485785,
1261
+ "grad_norm": 3.8504223823547363,
1262
+ "learning_rate": 0.00013756556618327678,
1263
+ "loss": 0.8964,
1264
+ "step": 4250
1265
+ },
1266
+ {
1267
+ "epoch": 0.9743034584923936,
1268
+ "grad_norm": 6.0986104011535645,
1269
+ "learning_rate": 0.00013717988275223697,
1270
+ "loss": 1.0507,
1271
+ "step": 4275
1272
+ },
1273
+ {
1274
+ "epoch": 0.9800011395362087,
1275
+ "grad_norm": 4.111655235290527,
1276
+ "learning_rate": 0.00013679419932119717,
1277
+ "loss": 0.8692,
1278
+ "step": 4300
1279
+ },
1280
+ {
1281
+ "epoch": 0.9856988205800239,
1282
+ "grad_norm": 6.2065958976745605,
1283
+ "learning_rate": 0.00013640851589015736,
1284
+ "loss": 1.0036,
1285
+ "step": 4325
1286
+ },
1287
+ {
1288
+ "epoch": 0.9913965016238391,
1289
+ "grad_norm": 2.76833176612854,
1290
+ "learning_rate": 0.00013602283245911756,
1291
+ "loss": 0.8781,
1292
+ "step": 4350
1293
+ },
1294
+ {
1295
+ "epoch": 0.9970941826676543,
1296
+ "grad_norm": 4.326218605041504,
1297
+ "learning_rate": 0.00013563714902807775,
1298
+ "loss": 0.8823,
1299
+ "step": 4375
1300
+ },
1301
+ {
1302
+ "epoch": 1.0029627941427839,
1303
+ "grad_norm": 3.9636833667755127,
1304
+ "learning_rate": 0.00013525146559703797,
1305
+ "loss": 0.8843,
1306
+ "step": 4400
1307
+ },
1308
+ {
1309
+ "epoch": 1.008660475186599,
1310
+ "grad_norm": 3.933739423751831,
1311
+ "learning_rate": 0.00013486578216599814,
1312
+ "loss": 0.7172,
1313
+ "step": 4425
1314
+ },
1315
+ {
1316
+ "epoch": 1.0143581562304143,
1317
+ "grad_norm": 3.0025434494018555,
1318
+ "learning_rate": 0.00013448009873495836,
1319
+ "loss": 0.7718,
1320
+ "step": 4450
1321
+ },
1322
+ {
1323
+ "epoch": 1.0200558372742294,
1324
+ "grad_norm": 3.4578757286071777,
1325
+ "learning_rate": 0.00013409441530391855,
1326
+ "loss": 0.7743,
1327
+ "step": 4475
1328
+ },
1329
+ {
1330
+ "epoch": 1.0257535183180446,
1331
+ "grad_norm": 2.7351579666137695,
1332
+ "learning_rate": 0.00013370873187287875,
1333
+ "loss": 0.7946,
1334
+ "step": 4500
1335
+ },
1336
+ {
1337
+ "epoch": 1.0314511993618598,
1338
+ "grad_norm": 2.814429759979248,
1339
+ "learning_rate": 0.00013332304844183894,
1340
+ "loss": 0.7994,
1341
+ "step": 4525
1342
+ },
1343
+ {
1344
+ "epoch": 1.037148880405675,
1345
+ "grad_norm": 2.7307941913604736,
1346
+ "learning_rate": 0.00013293736501079914,
1347
+ "loss": 0.7646,
1348
+ "step": 4550
1349
+ },
1350
+ {
1351
+ "epoch": 1.04284656144949,
1352
+ "grad_norm": 2.451763153076172,
1353
+ "learning_rate": 0.00013255168157975933,
1354
+ "loss": 0.8199,
1355
+ "step": 4575
1356
+ },
1357
+ {
1358
+ "epoch": 1.0485442424933051,
1359
+ "grad_norm": 2.954774856567383,
1360
+ "learning_rate": 0.00013216599814871953,
1361
+ "loss": 0.7679,
1362
+ "step": 4600
1363
+ },
1364
+ {
1365
+ "epoch": 1.0542419235371203,
1366
+ "grad_norm": 3.226649045944214,
1367
+ "learning_rate": 0.00013178031471767972,
1368
+ "loss": 0.7455,
1369
+ "step": 4625
1370
+ },
1371
+ {
1372
+ "epoch": 1.0599396045809355,
1373
+ "grad_norm": 4.065032958984375,
1374
+ "learning_rate": 0.00013139463128663994,
1375
+ "loss": 0.7711,
1376
+ "step": 4650
1377
+ },
1378
+ {
1379
+ "epoch": 1.0656372856247507,
1380
+ "grad_norm": 3.428072929382324,
1381
+ "learning_rate": 0.00013100894785560014,
1382
+ "loss": 0.7736,
1383
+ "step": 4675
1384
+ },
1385
+ {
1386
+ "epoch": 1.0713349666685659,
1387
+ "grad_norm": 5.30925989151001,
1388
+ "learning_rate": 0.00013062326442456033,
1389
+ "loss": 0.8166,
1390
+ "step": 4700
1391
+ },
1392
+ {
1393
+ "epoch": 1.077032647712381,
1394
+ "grad_norm": 4.497318744659424,
1395
+ "learning_rate": 0.00013023758099352053,
1396
+ "loss": 0.7342,
1397
+ "step": 4725
1398
+ },
1399
+ {
1400
+ "epoch": 1.0827303287561962,
1401
+ "grad_norm": 3.6386780738830566,
1402
+ "learning_rate": 0.00012985189756248072,
1403
+ "loss": 0.6903,
1404
+ "step": 4750
1405
+ },
1406
+ {
1407
+ "epoch": 1.0884280098000114,
1408
+ "grad_norm": 3.1318976879119873,
1409
+ "learning_rate": 0.00012946621413144091,
1410
+ "loss": 0.7043,
1411
+ "step": 4775
1412
+ },
1413
+ {
1414
+ "epoch": 1.0941256908438266,
1415
+ "grad_norm": 2.895879030227661,
1416
+ "learning_rate": 0.0001290805307004011,
1417
+ "loss": 0.7902,
1418
+ "step": 4800
1419
+ },
1420
+ {
1421
+ "epoch": 1.0998233718876418,
1422
+ "grad_norm": 2.9005753993988037,
1423
+ "learning_rate": 0.00012869484726936133,
1424
+ "loss": 0.8551,
1425
+ "step": 4825
1426
+ },
1427
+ {
1428
+ "epoch": 1.105521052931457,
1429
+ "grad_norm": 2.829397201538086,
1430
+ "learning_rate": 0.00012830916383832153,
1431
+ "loss": 0.7451,
1432
+ "step": 4850
1433
+ },
1434
+ {
1435
+ "epoch": 1.1112187339752722,
1436
+ "grad_norm": 6.609791278839111,
1437
+ "learning_rate": 0.0001279234804072817,
1438
+ "loss": 0.8076,
1439
+ "step": 4875
1440
+ },
1441
+ {
1442
+ "epoch": 1.1169164150190873,
1443
+ "grad_norm": 3.751443862915039,
1444
+ "learning_rate": 0.00012753779697624191,
1445
+ "loss": 0.7455,
1446
+ "step": 4900
1447
+ },
1448
+ {
1449
+ "epoch": 1.1226140960629023,
1450
+ "grad_norm": 3.541534900665283,
1451
+ "learning_rate": 0.0001271521135452021,
1452
+ "loss": 0.7613,
1453
+ "step": 4925
1454
+ },
1455
+ {
1456
+ "epoch": 1.1283117771067175,
1457
+ "grad_norm": 3.265768051147461,
1458
+ "learning_rate": 0.0001267664301141623,
1459
+ "loss": 0.6905,
1460
+ "step": 4950
1461
+ },
1462
+ {
1463
+ "epoch": 1.1340094581505327,
1464
+ "grad_norm": 3.051398992538452,
1465
+ "learning_rate": 0.0001263807466831225,
1466
+ "loss": 0.7622,
1467
+ "step": 4975
1468
+ },
1469
+ {
1470
+ "epoch": 1.1397071391943479,
1471
+ "grad_norm": 2.7481675148010254,
1472
+ "learning_rate": 0.0001259950632520827,
1473
+ "loss": 0.7318,
1474
+ "step": 5000
1475
+ },
1476
+ {
1477
+ "epoch": 1.1397071391943479,
1478
+ "eval_cer": 6.633806520310162,
1479
+ "eval_cer_en": 3.1808756211119373,
1480
+ "eval_cer_si": 8.259033317691225,
1481
+ "eval_cer_ta": 8.46151062212732,
1482
+ "eval_loss": 0.2541692852973938,
1483
+ "eval_runtime": 114.5705,
1484
+ "eval_samples_per_second": 6.546,
1485
+ "eval_steps_per_second": 0.82,
1486
+ "eval_wer": 24.71112099854156,
1487
+ "eval_wer_en": 5.169561621174525,
1488
+ "eval_wer_si": 39.015496809480396,
1489
+ "eval_wer_ta": 29.948304564969767,
1490
+ "step": 5000
1491
+ },
1492
+ {
1493
+ "epoch": 1.145404820238163,
1494
+ "grad_norm": 3.306461811065674,
1495
+ "learning_rate": 0.00012560937982104289,
1496
+ "loss": 0.7386,
1497
+ "step": 5025
1498
+ },
1499
+ {
1500
+ "epoch": 1.1511025012819782,
1501
+ "grad_norm": 2.8205950260162354,
1502
+ "learning_rate": 0.00012522369639000308,
1503
+ "loss": 0.7743,
1504
+ "step": 5050
1505
+ },
1506
+ {
1507
+ "epoch": 1.1568001823257934,
1508
+ "grad_norm": 2.839855194091797,
1509
+ "learning_rate": 0.0001248380129589633,
1510
+ "loss": 0.7688,
1511
+ "step": 5075
1512
+ },
1513
+ {
1514
+ "epoch": 1.1624978633696086,
1515
+ "grad_norm": 4.089432239532471,
1516
+ "learning_rate": 0.0001244523295279235,
1517
+ "loss": 0.7045,
1518
+ "step": 5100
1519
+ },
1520
+ {
1521
+ "epoch": 1.1681955444134238,
1522
+ "grad_norm": 4.832510471343994,
1523
+ "learning_rate": 0.00012406664609688366,
1524
+ "loss": 0.7877,
1525
+ "step": 5125
1526
+ },
1527
+ {
1528
+ "epoch": 1.173893225457239,
1529
+ "grad_norm": 3.631303071975708,
1530
+ "learning_rate": 0.00012368096266584389,
1531
+ "loss": 0.7407,
1532
+ "step": 5150
1533
+ },
1534
+ {
1535
+ "epoch": 1.1795909065010541,
1536
+ "grad_norm": 3.844489574432373,
1537
+ "learning_rate": 0.00012329527923480408,
1538
+ "loss": 0.7625,
1539
+ "step": 5175
1540
+ },
1541
+ {
1542
+ "epoch": 1.1852885875448693,
1543
+ "grad_norm": 2.8220715522766113,
1544
+ "learning_rate": 0.00012290959580376427,
1545
+ "loss": 0.7678,
1546
+ "step": 5200
1547
+ },
1548
+ {
1549
+ "epoch": 1.1909862685886843,
1550
+ "grad_norm": 3.201169729232788,
1551
+ "learning_rate": 0.00012252391237272447,
1552
+ "loss": 0.7628,
1553
+ "step": 5225
1554
+ },
1555
+ {
1556
+ "epoch": 1.1966839496324995,
1557
+ "grad_norm": 3.4348971843719482,
1558
+ "learning_rate": 0.00012213822894168466,
1559
+ "loss": 0.739,
1560
+ "step": 5250
1561
+ },
1562
+ {
1563
+ "epoch": 1.2023816306763146,
1564
+ "grad_norm": 4.422396183013916,
1565
+ "learning_rate": 0.00012175254551064487,
1566
+ "loss": 0.7389,
1567
+ "step": 5275
1568
+ },
1569
+ {
1570
+ "epoch": 1.2080793117201298,
1571
+ "grad_norm": 4.038496971130371,
1572
+ "learning_rate": 0.00012136686207960507,
1573
+ "loss": 0.7886,
1574
+ "step": 5300
1575
+ },
1576
+ {
1577
+ "epoch": 1.213776992763945,
1578
+ "grad_norm": 3.0748658180236816,
1579
+ "learning_rate": 0.00012098117864856526,
1580
+ "loss": 0.6752,
1581
+ "step": 5325
1582
+ },
1583
+ {
1584
+ "epoch": 1.2194746738077602,
1585
+ "grad_norm": 2.6235134601593018,
1586
+ "learning_rate": 0.00012059549521752547,
1587
+ "loss": 0.7128,
1588
+ "step": 5350
1589
+ },
1590
+ {
1591
+ "epoch": 1.2251723548515754,
1592
+ "grad_norm": 3.242774724960327,
1593
+ "learning_rate": 0.00012020981178648565,
1594
+ "loss": 0.7847,
1595
+ "step": 5375
1596
+ },
1597
+ {
1598
+ "epoch": 1.2308700358953906,
1599
+ "grad_norm": 6.7025532722473145,
1600
+ "learning_rate": 0.00011982412835544586,
1601
+ "loss": 0.7949,
1602
+ "step": 5400
1603
+ },
1604
+ {
1605
+ "epoch": 1.2365677169392058,
1606
+ "grad_norm": 3.9358043670654297,
1607
+ "learning_rate": 0.00011943844492440605,
1608
+ "loss": 0.7165,
1609
+ "step": 5425
1610
+ },
1611
+ {
1612
+ "epoch": 1.242265397983021,
1613
+ "grad_norm": 5.065573215484619,
1614
+ "learning_rate": 0.00011905276149336626,
1615
+ "loss": 0.7039,
1616
+ "step": 5450
1617
+ },
1618
+ {
1619
+ "epoch": 1.2479630790268361,
1620
+ "grad_norm": 4.109312057495117,
1621
+ "learning_rate": 0.00011866707806232644,
1622
+ "loss": 0.6779,
1623
+ "step": 5475
1624
+ },
1625
+ {
1626
+ "epoch": 1.2536607600706513,
1627
+ "grad_norm": 2.406324863433838,
1628
+ "learning_rate": 0.00011828139463128665,
1629
+ "loss": 0.7555,
1630
+ "step": 5500
1631
+ },
1632
+ {
1633
+ "epoch": 1.2593584411144665,
1634
+ "grad_norm": 2.9279966354370117,
1635
+ "learning_rate": 0.00011789571120024686,
1636
+ "loss": 0.6856,
1637
+ "step": 5525
1638
+ },
1639
+ {
1640
+ "epoch": 1.2650561221582817,
1641
+ "grad_norm": 2.020982265472412,
1642
+ "learning_rate": 0.00011751002776920704,
1643
+ "loss": 0.728,
1644
+ "step": 5550
1645
+ },
1646
+ {
1647
+ "epoch": 1.2707538032020969,
1648
+ "grad_norm": 6.182589530944824,
1649
+ "learning_rate": 0.00011712434433816723,
1650
+ "loss": 0.7684,
1651
+ "step": 5575
1652
+ },
1653
+ {
1654
+ "epoch": 1.276451484245912,
1655
+ "grad_norm": 9.388721466064453,
1656
+ "learning_rate": 0.00011673866090712744,
1657
+ "loss": 0.7304,
1658
+ "step": 5600
1659
+ },
1660
+ {
1661
+ "epoch": 1.282149165289727,
1662
+ "grad_norm": 4.756442546844482,
1663
+ "learning_rate": 0.00011635297747608765,
1664
+ "loss": 0.7081,
1665
+ "step": 5625
1666
+ },
1667
+ {
1668
+ "epoch": 1.2878468463335422,
1669
+ "grad_norm": 6.243655681610107,
1670
+ "learning_rate": 0.00011596729404504783,
1671
+ "loss": 0.7818,
1672
+ "step": 5650
1673
+ },
1674
+ {
1675
+ "epoch": 1.2935445273773574,
1676
+ "grad_norm": 3.19335675239563,
1677
+ "learning_rate": 0.00011558161061400802,
1678
+ "loss": 0.7524,
1679
+ "step": 5675
1680
+ },
1681
+ {
1682
+ "epoch": 1.2992422084211726,
1683
+ "grad_norm": 4.6336236000061035,
1684
+ "learning_rate": 0.00011521135452020981,
1685
+ "loss": 0.6826,
1686
+ "step": 5700
1687
+ },
1688
+ {
1689
+ "epoch": 1.3049398894649877,
1690
+ "grad_norm": 4.0369720458984375,
1691
+ "learning_rate": 0.00011482567108917002,
1692
+ "loss": 0.7428,
1693
+ "step": 5725
1694
+ },
1695
+ {
1696
+ "epoch": 1.310637570508803,
1697
+ "grad_norm": 3.9423370361328125,
1698
+ "learning_rate": 0.00011443998765813023,
1699
+ "loss": 0.7235,
1700
+ "step": 5750
1701
+ },
1702
+ {
1703
+ "epoch": 1.316335251552618,
1704
+ "grad_norm": 2.53438401222229,
1705
+ "learning_rate": 0.00011405430422709041,
1706
+ "loss": 0.656,
1707
+ "step": 5775
1708
+ },
1709
+ {
1710
+ "epoch": 1.3220329325964333,
1711
+ "grad_norm": 3.988180160522461,
1712
+ "learning_rate": 0.0001136686207960506,
1713
+ "loss": 0.7299,
1714
+ "step": 5800
1715
+ },
1716
+ {
1717
+ "epoch": 1.3277306136402485,
1718
+ "grad_norm": 6.210958480834961,
1719
+ "learning_rate": 0.00011328293736501081,
1720
+ "loss": 0.717,
1721
+ "step": 5825
1722
+ },
1723
+ {
1724
+ "epoch": 1.3334282946840637,
1725
+ "grad_norm": 3.7127974033355713,
1726
+ "learning_rate": 0.00011289725393397099,
1727
+ "loss": 0.7563,
1728
+ "step": 5850
1729
+ },
1730
+ {
1731
+ "epoch": 1.3391259757278786,
1732
+ "grad_norm": 3.8320415019989014,
1733
+ "learning_rate": 0.0001125115705029312,
1734
+ "loss": 0.7699,
1735
+ "step": 5875
1736
+ },
1737
+ {
1738
+ "epoch": 1.3448236567716938,
1739
+ "grad_norm": 2.8862693309783936,
1740
+ "learning_rate": 0.0001121258870718914,
1741
+ "loss": 0.6879,
1742
+ "step": 5900
1743
+ },
1744
+ {
1745
+ "epoch": 1.350521337815509,
1746
+ "grad_norm": 3.1378445625305176,
1747
+ "learning_rate": 0.0001117402036408516,
1748
+ "loss": 0.696,
1749
+ "step": 5925
1750
+ },
1751
+ {
1752
+ "epoch": 1.3562190188593242,
1753
+ "grad_norm": 3.703091859817505,
1754
+ "learning_rate": 0.00011135452020981178,
1755
+ "loss": 0.7181,
1756
+ "step": 5950
1757
+ },
1758
+ {
1759
+ "epoch": 1.3619166999031393,
1760
+ "grad_norm": 4.001578330993652,
1761
+ "learning_rate": 0.00011096883677877199,
1762
+ "loss": 0.7389,
1763
+ "step": 5975
1764
+ },
1765
+ {
1766
+ "epoch": 1.3676143809469545,
1767
+ "grad_norm": 4.707931041717529,
1768
+ "learning_rate": 0.0001105831533477322,
1769
+ "loss": 0.7509,
1770
+ "step": 6000
1771
+ },
1772
+ {
1773
+ "epoch": 1.3676143809469545,
1774
+ "eval_cer": 6.268446254831395,
1775
+ "eval_cer_en": 3.1847881372510662,
1776
+ "eval_cer_si": 7.711559518223056,
1777
+ "eval_cer_ta": 7.908991109020063,
1778
+ "eval_loss": 0.2612861394882202,
1779
+ "eval_runtime": 114.905,
1780
+ "eval_samples_per_second": 6.527,
1781
+ "eval_steps_per_second": 0.818,
1782
+ "eval_wer": 22.95756267244273,
1783
+ "eval_wer_en": 5.086848635235732,
1784
+ "eval_wer_si": 35.186873290793066,
1785
+ "eval_wer_ta": 28.598966091299395,
1786
+ "step": 6000
1787
  }
1788
  ],
1789
  "logging_steps": 25,
 
1798
  "should_evaluate": false,
1799
  "should_log": false,
1800
  "should_save": true,
1801
+ "should_training_stop": false
1802
  },
1803
  "attributes": {}
1804
  }
1805
  },
1806
+ "total_flos": 1.027722952802304e+20,
1807
  "train_batch_size": 4,
1808
  "trial_name": null,
1809
  "trial_params": null