Training in progress, step 5000, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -34,12 +34,12 @@
|
|
| 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,
|
| 45 |
"task_type": null,
|
|
|
|
| 34 |
"rank_pattern": {},
|
| 35 |
"revision": null,
|
| 36 |
"target_modules": [
|
|
|
|
| 37 |
"fc1",
|
| 38 |
"fc2",
|
| 39 |
+
"k_proj",
|
| 40 |
+
"out_proj",
|
| 41 |
"v_proj",
|
| 42 |
+
"q_proj"
|
|
|
|
| 43 |
],
|
| 44 |
"target_parameters": null,
|
| 45 |
"task_type": null,
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 138520208
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9dac2facb076a10be1d1c6a58e123469c5a8143251c5f3da45b517fc823f01b
|
| 3 |
size 138520208
|
last-checkpoint/optimizer.pt
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 sha256:044b3d93b445073511b70983fe4c36a9dcc45f003be5fb72334d4aa063b344b5
|
| 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:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:925943a30664dfa9ac48d0cfc92698be8d8ccc7882d0a78408688c3285176180
|
| 3 |
size 14645
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:00359e2973671a14feb64a06f8f8c678e221e9b494b0b83bfe195ff21326d0c8
|
| 3 |
size 1383
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c4effe53d359d85d520fceb9a42bf463905c991a3f1fdf539171658e89baf78
|
| 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":
|
| 6 |
"eval_steps": 1000,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -1078,6 +1078,416 @@
|
|
| 1078 |
"learning_rate": 0.00014679111385374884,
|
| 1079 |
"loss": 0.1867,
|
| 1080 |
"step": 3650
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1081 |
}
|
| 1082 |
],
|
| 1083 |
"logging_steps": 25,
|
|
@@ -1092,12 +1502,12 @@
|
|
| 1092 |
"should_evaluate": false,
|
| 1093 |
"should_log": false,
|
| 1094 |
"should_save": true,
|
| 1095 |
-
"should_training_stop":
|
| 1096 |
},
|
| 1097 |
"attributes": {}
|
| 1098 |
}
|
| 1099 |
},
|
| 1100 |
-
"total_flos":
|
| 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": 1.1397071391943479,
|
| 6 |
"eval_steps": 1000,
|
| 7 |
+
"global_step": 5000,
|
| 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": 0.5877,
|
| 1087 |
+
"step": 3675
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.8432567944846447,
|
| 1091 |
+
"grad_norm": 2.234744071960449,
|
| 1092 |
+
"learning_rate": 0.00014605060166615242,
|
| 1093 |
+
"loss": 0.6687,
|
| 1094 |
+
"step": 3700
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.8489544755284599,
|
| 1098 |
+
"grad_norm": 3.9890682697296143,
|
| 1099 |
+
"learning_rate": 0.00014566491823511264,
|
| 1100 |
+
"loss": 1.2452,
|
| 1101 |
+
"step": 3725
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.8546521565722751,
|
| 1105 |
+
"grad_norm": 4.3173089027404785,
|
| 1106 |
+
"learning_rate": 0.0001452792348040728,
|
| 1107 |
+
"loss": 1.0955,
|
| 1108 |
+
"step": 3750
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.8603498376160903,
|
| 1112 |
+
"grad_norm": 5.6849870681762695,
|
| 1113 |
+
"learning_rate": 0.00014489355137303303,
|
| 1114 |
+
"loss": 1.077,
|
| 1115 |
+
"step": 3775
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.8660475186599054,
|
| 1119 |
+
"grad_norm": 3.146728992462158,
|
| 1120 |
+
"learning_rate": 0.00014450786794199322,
|
| 1121 |
+
"loss": 1.1362,
|
| 1122 |
+
"step": 3800
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8717451997037206,
|
| 1126 |
+
"grad_norm": 4.720071315765381,
|
| 1127 |
+
"learning_rate": 0.00014412218451095342,
|
| 1128 |
+
"loss": 0.9808,
|
| 1129 |
+
"step": 3825
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.8774428807475357,
|
| 1133 |
+
"grad_norm": 4.590244293212891,
|
| 1134 |
+
"learning_rate": 0.0001437365010799136,
|
| 1135 |
+
"loss": 1.0131,
|
| 1136 |
+
"step": 3850
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.8831405617913509,
|
| 1140 |
+
"grad_norm": 3.2722008228302,
|
| 1141 |
+
"learning_rate": 0.0001433508176488738,
|
| 1142 |
+
"loss": 1.0495,
|
| 1143 |
+
"step": 3875
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.8888382428351661,
|
| 1147 |
+
"grad_norm": 3.293778657913208,
|
| 1148 |
+
"learning_rate": 0.00014296513421783403,
|
| 1149 |
+
"loss": 1.0056,
|
| 1150 |
+
"step": 3900
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.8945359238789813,
|
| 1154 |
+
"grad_norm": NaN,
|
| 1155 |
+
"learning_rate": 0.0001425794507867942,
|
| 1156 |
+
"loss": 1.0383,
|
| 1157 |
+
"step": 3925
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.9002336049227965,
|
| 1161 |
+
"grad_norm": 3.6257212162017822,
|
| 1162 |
+
"learning_rate": 0.000142209194692996,
|
| 1163 |
+
"loss": 0.9983,
|
| 1164 |
+
"step": 3950
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.9059312859666115,
|
| 1168 |
+
"grad_norm": 3.990129232406616,
|
| 1169 |
+
"learning_rate": 0.00014182351126195618,
|
| 1170 |
+
"loss": 0.9433,
|
| 1171 |
+
"step": 3975
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.9116289670104267,
|
| 1175 |
+
"grad_norm": 4.548169136047363,
|
| 1176 |
+
"learning_rate": 0.0001414378278309164,
|
| 1177 |
+
"loss": 1.0048,
|
| 1178 |
+
"step": 4000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.9116289670104267,
|
| 1182 |
+
"eval_cer": 7.389038091583553,
|
| 1183 |
+
"eval_cer_en": 3.0478500723815487,
|
| 1184 |
+
"eval_cer_si": 10.402002189895198,
|
| 1185 |
+
"eval_cer_ta": 8.717262012473913,
|
| 1186 |
+
"eval_loss": 0.26376235485076904,
|
| 1187 |
+
"eval_runtime": 120.5467,
|
| 1188 |
+
"eval_samples_per_second": 6.222,
|
| 1189 |
+
"eval_steps_per_second": 0.78,
|
| 1190 |
+
"eval_wer": 27.785687057864198,
|
| 1191 |
+
"eval_wer_en": 4.983457402812242,
|
| 1192 |
+
"eval_wer_si": 47.1285323609845,
|
| 1193 |
+
"eval_wer_ta": 31.245071409795848,
|
| 1194 |
+
"step": 4000
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.9173266480542419,
|
| 1198 |
+
"grad_norm": 3.8549602031707764,
|
| 1199 |
+
"learning_rate": 0.0001410521443998766,
|
| 1200 |
+
"loss": 1.0129,
|
| 1201 |
+
"step": 4025
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.9230243290980571,
|
| 1205 |
+
"grad_norm": 3.753631353378296,
|
| 1206 |
+
"learning_rate": 0.0001406664609688368,
|
| 1207 |
+
"loss": 1.0125,
|
| 1208 |
+
"step": 4050
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.9287220101418723,
|
| 1212 |
+
"grad_norm": 5.59144401550293,
|
| 1213 |
+
"learning_rate": 0.00014028077753779698,
|
| 1214 |
+
"loss": 0.9451,
|
| 1215 |
+
"step": 4075
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 0.9344196911856875,
|
| 1219 |
+
"grad_norm": 2.9432716369628906,
|
| 1220 |
+
"learning_rate": 0.00013989509410675718,
|
| 1221 |
+
"loss": 0.9489,
|
| 1222 |
+
"step": 4100
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 0.9401173722295026,
|
| 1226 |
+
"grad_norm": 5.580732345581055,
|
| 1227 |
+
"learning_rate": 0.00013950941067571737,
|
| 1228 |
+
"loss": 1.0758,
|
| 1229 |
+
"step": 4125
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 0.9458150532733177,
|
| 1233 |
+
"grad_norm": 3.4677042961120605,
|
| 1234 |
+
"learning_rate": 0.00013912372724467757,
|
| 1235 |
+
"loss": 0.9962,
|
| 1236 |
+
"step": 4150
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 0.9515127343171329,
|
| 1240 |
+
"grad_norm": 4.347315788269043,
|
| 1241 |
+
"learning_rate": 0.00013873804381363776,
|
| 1242 |
+
"loss": 0.9171,
|
| 1243 |
+
"step": 4175
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 0.9572104153609481,
|
| 1247 |
+
"grad_norm": 6.545660972595215,
|
| 1248 |
+
"learning_rate": 0.00013835236038259798,
|
| 1249 |
+
"loss": 0.9835,
|
| 1250 |
+
"step": 4200
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.9629080964047633,
|
| 1254 |
+
"grad_norm": 5.0610222816467285,
|
| 1255 |
+
"learning_rate": 0.00013796667695155815,
|
| 1256 |
+
"loss": 0.9005,
|
| 1257 |
+
"step": 4225
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 0.9686057774485785,
|
| 1261 |
+
"grad_norm": 3.217003107070923,
|
| 1262 |
+
"learning_rate": 0.00013758099352051837,
|
| 1263 |
+
"loss": 0.9004,
|
| 1264 |
+
"step": 4250
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 0.9743034584923936,
|
| 1268 |
+
"grad_norm": 5.750592231750488,
|
| 1269 |
+
"learning_rate": 0.00013719531008947857,
|
| 1270 |
+
"loss": 1.0545,
|
| 1271 |
+
"step": 4275
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 0.9800011395362087,
|
| 1275 |
+
"grad_norm": 4.246372699737549,
|
| 1276 |
+
"learning_rate": 0.00013680962665843876,
|
| 1277 |
+
"loss": 0.8752,
|
| 1278 |
+
"step": 4300
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 0.9856988205800239,
|
| 1282 |
+
"grad_norm": 4.3279128074646,
|
| 1283 |
+
"learning_rate": 0.00013642394322739896,
|
| 1284 |
+
"loss": 0.9912,
|
| 1285 |
+
"step": 4325
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.9913965016238391,
|
| 1289 |
+
"grad_norm": 2.8706257343292236,
|
| 1290 |
+
"learning_rate": 0.00013603825979635915,
|
| 1291 |
+
"loss": 0.8692,
|
| 1292 |
+
"step": 4350
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.9970941826676543,
|
| 1296 |
+
"grad_norm": 4.598703861236572,
|
| 1297 |
+
"learning_rate": 0.00013565257636531937,
|
| 1298 |
+
"loss": 0.8828,
|
| 1299 |
+
"step": 4375
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 1.0029627941427839,
|
| 1303 |
+
"grad_norm": 3.949667453765869,
|
| 1304 |
+
"learning_rate": 0.00013526689293427954,
|
| 1305 |
+
"loss": 0.8835,
|
| 1306 |
+
"step": 4400
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 1.008660475186599,
|
| 1310 |
+
"grad_norm": 3.27917742729187,
|
| 1311 |
+
"learning_rate": 0.00013488120950323976,
|
| 1312 |
+
"loss": 0.704,
|
| 1313 |
+
"step": 4425
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 1.0143581562304143,
|
| 1317 |
+
"grad_norm": 3.343886613845825,
|
| 1318 |
+
"learning_rate": 0.00013449552607219995,
|
| 1319 |
+
"loss": 0.776,
|
| 1320 |
+
"step": 4450
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.0200558372742294,
|
| 1324 |
+
"grad_norm": 3.5439653396606445,
|
| 1325 |
+
"learning_rate": 0.00013410984264116012,
|
| 1326 |
+
"loss": 0.7893,
|
| 1327 |
+
"step": 4475
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 1.0257535183180446,
|
| 1331 |
+
"grad_norm": 4.269862651824951,
|
| 1332 |
+
"learning_rate": 0.00013372415921012034,
|
| 1333 |
+
"loss": 0.7819,
|
| 1334 |
+
"step": 4500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 1.0314511993618598,
|
| 1338 |
+
"grad_norm": 2.7746996879577637,
|
| 1339 |
+
"learning_rate": 0.00013333847577908054,
|
| 1340 |
+
"loss": 0.7935,
|
| 1341 |
+
"step": 4525
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 1.037148880405675,
|
| 1345 |
+
"grad_norm": 2.3175647258758545,
|
| 1346 |
+
"learning_rate": 0.00013295279234804073,
|
| 1347 |
+
"loss": 0.7833,
|
| 1348 |
+
"step": 4550
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 1.04284656144949,
|
| 1352 |
+
"grad_norm": 2.416553020477295,
|
| 1353 |
+
"learning_rate": 0.00013256710891700093,
|
| 1354 |
+
"loss": 0.8198,
|
| 1355 |
+
"step": 4575
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 1.0485442424933051,
|
| 1359 |
+
"grad_norm": 3.0306460857391357,
|
| 1360 |
+
"learning_rate": 0.00013218142548596112,
|
| 1361 |
+
"loss": 0.7602,
|
| 1362 |
+
"step": 4600
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 1.0542419235371203,
|
| 1366 |
+
"grad_norm": 2.776203155517578,
|
| 1367 |
+
"learning_rate": 0.00013179574205492134,
|
| 1368 |
+
"loss": 0.7477,
|
| 1369 |
+
"step": 4625
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 1.0599396045809355,
|
| 1373 |
+
"grad_norm": 3.6620233058929443,
|
| 1374 |
+
"learning_rate": 0.0001314100586238815,
|
| 1375 |
+
"loss": 0.7726,
|
| 1376 |
+
"step": 4650
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 1.0656372856247507,
|
| 1380 |
+
"grad_norm": 2.7935216426849365,
|
| 1381 |
+
"learning_rate": 0.00013102437519284173,
|
| 1382 |
+
"loss": 0.767,
|
| 1383 |
+
"step": 4675
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 1.0713349666685659,
|
| 1387 |
+
"grad_norm": 4.359261512756348,
|
| 1388 |
+
"learning_rate": 0.00013063869176180193,
|
| 1389 |
+
"loss": 0.8269,
|
| 1390 |
+
"step": 4700
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 1.077032647712381,
|
| 1394 |
+
"grad_norm": 3.0299580097198486,
|
| 1395 |
+
"learning_rate": 0.00013025300833076212,
|
| 1396 |
+
"loss": 0.7262,
|
| 1397 |
+
"step": 4725
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 1.0827303287561962,
|
| 1401 |
+
"grad_norm": 3.7441720962524414,
|
| 1402 |
+
"learning_rate": 0.00012986732489972231,
|
| 1403 |
+
"loss": 0.685,
|
| 1404 |
+
"step": 4750
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 1.0884280098000114,
|
| 1408 |
+
"grad_norm": 3.534374475479126,
|
| 1409 |
+
"learning_rate": 0.0001294816414686825,
|
| 1410 |
+
"loss": 0.702,
|
| 1411 |
+
"step": 4775
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 1.0941256908438266,
|
| 1415 |
+
"grad_norm": 2.7458009719848633,
|
| 1416 |
+
"learning_rate": 0.0001290959580376427,
|
| 1417 |
+
"loss": 0.808,
|
| 1418 |
+
"step": 4800
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 1.0998233718876418,
|
| 1422 |
+
"grad_norm": 2.5479207038879395,
|
| 1423 |
+
"learning_rate": 0.0001287102746066029,
|
| 1424 |
+
"loss": 0.8506,
|
| 1425 |
+
"step": 4825
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.105521052931457,
|
| 1429 |
+
"grad_norm": 3.05680775642395,
|
| 1430 |
+
"learning_rate": 0.0001283245911755631,
|
| 1431 |
+
"loss": 0.7608,
|
| 1432 |
+
"step": 4850
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.1112187339752722,
|
| 1436 |
+
"grad_norm": 39.71477508544922,
|
| 1437 |
+
"learning_rate": 0.00012793890774452331,
|
| 1438 |
+
"loss": 0.7914,
|
| 1439 |
+
"step": 4875
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.1169164150190873,
|
| 1443 |
+
"grad_norm": 3.3444302082061768,
|
| 1444 |
+
"learning_rate": 0.00012755322431348348,
|
| 1445 |
+
"loss": 0.75,
|
| 1446 |
+
"step": 4900
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 1.1226140960629023,
|
| 1450 |
+
"grad_norm": 3.3730239868164062,
|
| 1451 |
+
"learning_rate": 0.0001271675408824437,
|
| 1452 |
+
"loss": 0.7536,
|
| 1453 |
+
"step": 4925
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 1.1283117771067175,
|
| 1457 |
+
"grad_norm": 3.1191494464874268,
|
| 1458 |
+
"learning_rate": 0.0001267818574514039,
|
| 1459 |
+
"loss": 0.707,
|
| 1460 |
+
"step": 4950
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 1.1340094581505327,
|
| 1464 |
+
"grad_norm": 3.301866054534912,
|
| 1465 |
+
"learning_rate": 0.0001263961740203641,
|
| 1466 |
+
"loss": 0.7679,
|
| 1467 |
+
"step": 4975
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 1.1397071391943479,
|
| 1471 |
+
"grad_norm": 3.590381145477295,
|
| 1472 |
+
"learning_rate": 0.00012601049058932429,
|
| 1473 |
+
"loss": 0.7362,
|
| 1474 |
+
"step": 5000
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 1.1397071391943479,
|
| 1478 |
+
"eval_cer": 6.585144399193683,
|
| 1479 |
+
"eval_cer_en": 3.08306271763371,
|
| 1480 |
+
"eval_cer_si": 8.227749100578759,
|
| 1481 |
+
"eval_cer_ta": 8.444621379368582,
|
| 1482 |
+
"eval_loss": 0.2518573999404907,
|
| 1483 |
+
"eval_runtime": 119.7953,
|
| 1484 |
+
"eval_samples_per_second": 6.261,
|
| 1485 |
+
"eval_steps_per_second": 0.785,
|
| 1486 |
+
"eval_wer": 24.63286980701743,
|
| 1487 |
+
"eval_wer_en": 4.962779156327544,
|
| 1488 |
+
"eval_wer_si": 39.197812215132174,
|
| 1489 |
+
"eval_wer_ta": 29.73801804959257,
|
| 1490 |
+
"step": 5000
|
| 1491 |
}
|
| 1492 |
],
|
| 1493 |
"logging_steps": 25,
|
|
|
|
| 1502 |
"should_evaluate": false,
|
| 1503 |
"should_log": false,
|
| 1504 |
"should_save": true,
|
| 1505 |
+
"should_training_stop": false
|
| 1506 |
},
|
| 1507 |
"attributes": {}
|
| 1508 |
}
|
| 1509 |
},
|
| 1510 |
+
"total_flos": 8.56453634555904e+19,
|
| 1511 |
"train_batch_size": 4,
|
| 1512 |
"trial_name": null,
|
| 1513 |
"trial_params": null
|