Check commited on
Commit
6b40a07
·
1 Parent(s): f621031

"auto-commit"

Browse files
Files changed (38) hide show
  1. model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/config.json +0 -0
  2. model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/optimizer.pt +1 -1
  3. model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/preprocessor_config.json +0 -0
  4. model-bin/finetune/base/{checkpoint-15183 → checkpoint-15804}/pytorch_model.bin +1 -1
  5. model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/rng_state.pth +1 -1
  6. model-bin/finetune/base/{checkpoint-15183 → checkpoint-15804}/scaler.pt +1 -1
  7. model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/scheduler.pt +1 -1
  8. model-bin/finetune/base/{checkpoint-15183 → checkpoint-15804}/trainer_state.json +3776 -5
  9. model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/training_args.bin +0 -0
  10. model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/config.json +0 -0
  11. model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/optimizer.pt +1 -1
  12. model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/preprocessor_config.json +0 -0
  13. model-bin/finetune/base/{checkpoint-14810 → checkpoint-16425}/pytorch_model.bin +1 -1
  14. model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/rng_state.pth +2 -2
  15. model-bin/finetune/base/{checkpoint-14810 → checkpoint-16425}/scaler.pt +1 -1
  16. model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/scheduler.pt +1 -1
  17. model-bin/finetune/base/{checkpoint-14810 → checkpoint-16425}/trainer_state.json +0 -0
  18. model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/training_args.bin +0 -0
  19. model-bin/finetune/base/log/1629515621.4037716/events.out.tfevents.1629515621.35c5ee475cfb.1049.43 +3 -0
  20. model-bin/finetune/base/log/1629516257.3050275/events.out.tfevents.1629516257.35c5ee475cfb.1049.45 +3 -0
  21. model-bin/finetune/base/log/1629516898.1683943/events.out.tfevents.1629516898.35c5ee475cfb.1049.47 +3 -0
  22. model-bin/finetune/base/log/1629517537.5125346/events.out.tfevents.1629517537.35c5ee475cfb.1049.49 +3 -0
  23. model-bin/finetune/base/log/1629518168.3557167/events.out.tfevents.1629518168.35c5ee475cfb.1049.51 +3 -0
  24. model-bin/finetune/base/log/1629518794.4714847/events.out.tfevents.1629518794.35c5ee475cfb.1049.53 +3 -0
  25. model-bin/finetune/base/log/1629519424.9971457/events.out.tfevents.1629519424.35c5ee475cfb.1049.55 +3 -0
  26. model-bin/finetune/base/log/1629520050.9530234/events.out.tfevents.1629520050.35c5ee475cfb.1049.57 +3 -0
  27. model-bin/finetune/base/log/1629520692.8460991/events.out.tfevents.1629520692.35c5ee475cfb.1049.59 +3 -0
  28. model-bin/finetune/base/log/1629521428.9057782/events.out.tfevents.1629521428.35c5ee475cfb.1049.61 +3 -0
  29. model-bin/finetune/base/log/events.out.tfevents.1629515621.35c5ee475cfb.1049.42 +3 -0
  30. model-bin/finetune/base/log/events.out.tfevents.1629516257.35c5ee475cfb.1049.44 +3 -0
  31. model-bin/finetune/base/log/events.out.tfevents.1629516898.35c5ee475cfb.1049.46 +3 -0
  32. model-bin/finetune/base/log/events.out.tfevents.1629517537.35c5ee475cfb.1049.48 +3 -0
  33. model-bin/finetune/base/log/events.out.tfevents.1629518168.35c5ee475cfb.1049.50 +3 -0
  34. model-bin/finetune/base/log/events.out.tfevents.1629518794.35c5ee475cfb.1049.52 +3 -0
  35. model-bin/finetune/base/log/events.out.tfevents.1629519424.35c5ee475cfb.1049.54 +3 -0
  36. model-bin/finetune/base/log/events.out.tfevents.1629520050.35c5ee475cfb.1049.56 +3 -0
  37. model-bin/finetune/base/log/events.out.tfevents.1629520692.35c5ee475cfb.1049.58 +3 -0
  38. model-bin/finetune/base/log/events.out.tfevents.1629521428.35c5ee475cfb.1049.60 +3 -0
model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/config.json RENAMED
File without changes
model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/optimizer.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b9003e3bb2f85e7e059cf8b234e0fb67aafb267d7936141a07dcce072513b63f
3
  size 722165009
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1267bf984c5759effbb4243352e9e38cda800ae54fa9233fe60af40da04c5b66
3
  size 722165009
model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/preprocessor_config.json RENAMED
File without changes
model-bin/finetune/base/{checkpoint-15183 → checkpoint-15804}/pytorch_model.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3040c71a62bc350d839a87f8e91c71a716fc00d96df7011ff50f6f052134607c
3
  size 377909911
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38bdaad9404abe5b2aa7ff8a0ac8aec2701cf52a4eee44e70aa326dcaf573ae4
3
  size 377909911
model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/rng_state.pth RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f12ff574e2074e68c73a5a192d4ff3b5b1a9338e9f0d95ac8ec15d9bc1e6a8d8
3
  size 14503
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd598dbe65be03b6ece4a6c58f568f3844df4f1f69b5426295526d9da1a8d9be
3
  size 14503
model-bin/finetune/base/{checkpoint-15183 → checkpoint-15804}/scaler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a5340902b17e35966e7e8d13c891b4a72edbb8cbd01bf429032179aa2bf58052
3
  size 559
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fff7c90dfcdd1e18ff8a802f53d8ab91342064d1a27f602a5b57345773e9dc21
3
  size 559
model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/scheduler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e4efc650f4e1e6bf4702acc9abf71067d68b5a46118a24332b51e04b8113a70a
3
  size 623
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a9df0cb8f8cf902a1007e3d887bcdeb3a9f0b7fe49bb010099f8ae899e3ece4
3
  size 623
model-bin/finetune/base/{checkpoint-15183 → checkpoint-15804}/trainer_state.json RENAMED
@@ -1,8 +1,8 @@
1
  {
2
- "best_metric": 0.2679527616812147,
3
- "best_model_checkpoint": "./model-bin/finetune/base/checkpoint-14810",
4
- "epoch": 122.0,
5
- "global_step": 15183,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
@@ -92202,11 +92202,3782 @@
92202
  "eval_steps_per_second": 0.674,
92203
  "eval_wer": 0.2688305978898007,
92204
  "step": 15183
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
92205
  }
92206
  ],
92207
  "max_steps": 620000,
92208
  "num_train_epochs": 5000,
92209
- "total_flos": 4.269612444847596e+19,
92210
  "trial_name": null,
92211
  "trial_params": null
92212
  }
 
1
  {
2
+ "best_metric": 0.25577642485146335,
3
+ "best_model_checkpoint": "./model-bin/finetune/base/checkpoint-15804",
4
+ "epoch": 126.99598393574297,
5
+ "global_step": 15804,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
 
92202
  "eval_steps_per_second": 0.674,
92203
  "eval_wer": 0.2688305978898007,
92204
  "step": 15183
92205
+ },
92206
+ {
92207
+ "epoch": 122.01,
92208
+ "learning_rate": 9.770953150242327e-06,
92209
+ "loss": 0.638,
92210
+ "step": 15184
92211
+ },
92212
+ {
92213
+ "epoch": 122.02,
92214
+ "learning_rate": 9.770936995153473e-06,
92215
+ "loss": 0.42,
92216
+ "step": 15185
92217
+ },
92218
+ {
92219
+ "epoch": 122.02,
92220
+ "learning_rate": 9.770920840064622e-06,
92221
+ "loss": 0.6889,
92222
+ "step": 15186
92223
+ },
92224
+ {
92225
+ "epoch": 122.03,
92226
+ "learning_rate": 9.770904684975769e-06,
92227
+ "loss": 0.5918,
92228
+ "step": 15187
92229
+ },
92230
+ {
92231
+ "epoch": 122.04,
92232
+ "learning_rate": 9.770888529886916e-06,
92233
+ "loss": 0.503,
92234
+ "step": 15188
92235
+ },
92236
+ {
92237
+ "epoch": 122.05,
92238
+ "learning_rate": 9.770872374798062e-06,
92239
+ "loss": 0.5364,
92240
+ "step": 15189
92241
+ },
92242
+ {
92243
+ "epoch": 122.06,
92244
+ "learning_rate": 9.77085621970921e-06,
92245
+ "loss": 0.523,
92246
+ "step": 15190
92247
+ },
92248
+ {
92249
+ "epoch": 122.06,
92250
+ "learning_rate": 9.770840064620356e-06,
92251
+ "loss": 0.5624,
92252
+ "step": 15191
92253
+ },
92254
+ {
92255
+ "epoch": 122.07,
92256
+ "learning_rate": 9.770823909531503e-06,
92257
+ "loss": 0.5685,
92258
+ "step": 15192
92259
+ },
92260
+ {
92261
+ "epoch": 122.08,
92262
+ "learning_rate": 9.77080775444265e-06,
92263
+ "loss": 0.5897,
92264
+ "step": 15193
92265
+ },
92266
+ {
92267
+ "epoch": 122.09,
92268
+ "learning_rate": 9.770791599353797e-06,
92269
+ "loss": 0.5451,
92270
+ "step": 15194
92271
+ },
92272
+ {
92273
+ "epoch": 122.1,
92274
+ "learning_rate": 9.770775444264943e-06,
92275
+ "loss": 0.4997,
92276
+ "step": 15195
92277
+ },
92278
+ {
92279
+ "epoch": 122.1,
92280
+ "learning_rate": 9.770759289176092e-06,
92281
+ "loss": 0.8023,
92282
+ "step": 15196
92283
+ },
92284
+ {
92285
+ "epoch": 122.11,
92286
+ "learning_rate": 9.770743134087239e-06,
92287
+ "loss": 0.5179,
92288
+ "step": 15197
92289
+ },
92290
+ {
92291
+ "epoch": 122.12,
92292
+ "learning_rate": 9.770726978998386e-06,
92293
+ "loss": 0.6488,
92294
+ "step": 15198
92295
+ },
92296
+ {
92297
+ "epoch": 122.13,
92298
+ "learning_rate": 9.770710823909532e-06,
92299
+ "loss": 0.615,
92300
+ "step": 15199
92301
+ },
92302
+ {
92303
+ "epoch": 122.14,
92304
+ "learning_rate": 9.77069466882068e-06,
92305
+ "loss": 1.0009,
92306
+ "step": 15200
92307
+ },
92308
+ {
92309
+ "epoch": 122.15,
92310
+ "learning_rate": 9.770678513731826e-06,
92311
+ "loss": 0.6498,
92312
+ "step": 15201
92313
+ },
92314
+ {
92315
+ "epoch": 122.15,
92316
+ "learning_rate": 9.770662358642975e-06,
92317
+ "loss": 0.7208,
92318
+ "step": 15202
92319
+ },
92320
+ {
92321
+ "epoch": 122.16,
92322
+ "learning_rate": 9.77064620355412e-06,
92323
+ "loss": 0.8116,
92324
+ "step": 15203
92325
+ },
92326
+ {
92327
+ "epoch": 122.17,
92328
+ "learning_rate": 9.770630048465267e-06,
92329
+ "loss": 1.0523,
92330
+ "step": 15204
92331
+ },
92332
+ {
92333
+ "epoch": 122.18,
92334
+ "learning_rate": 9.770613893376413e-06,
92335
+ "loss": 1.3636,
92336
+ "step": 15205
92337
+ },
92338
+ {
92339
+ "epoch": 122.19,
92340
+ "learning_rate": 9.770597738287562e-06,
92341
+ "loss": 1.3104,
92342
+ "step": 15206
92343
+ },
92344
+ {
92345
+ "epoch": 122.19,
92346
+ "learning_rate": 9.770581583198709e-06,
92347
+ "loss": 1.9627,
92348
+ "step": 15207
92349
+ },
92350
+ {
92351
+ "epoch": 122.2,
92352
+ "learning_rate": 9.770565428109856e-06,
92353
+ "loss": 2.8652,
92354
+ "step": 15208
92355
+ },
92356
+ {
92357
+ "epoch": 122.21,
92358
+ "learning_rate": 9.770549273021002e-06,
92359
+ "loss": 0.7252,
92360
+ "step": 15209
92361
+ },
92362
+ {
92363
+ "epoch": 122.22,
92364
+ "learning_rate": 9.77053311793215e-06,
92365
+ "loss": 0.5633,
92366
+ "step": 15210
92367
+ },
92368
+ {
92369
+ "epoch": 122.23,
92370
+ "learning_rate": 9.770516962843296e-06,
92371
+ "loss": 0.6688,
92372
+ "step": 15211
92373
+ },
92374
+ {
92375
+ "epoch": 122.23,
92376
+ "learning_rate": 9.770500807754445e-06,
92377
+ "loss": 0.4263,
92378
+ "step": 15212
92379
+ },
92380
+ {
92381
+ "epoch": 122.24,
92382
+ "learning_rate": 9.77048465266559e-06,
92383
+ "loss": 1.1555,
92384
+ "step": 15213
92385
+ },
92386
+ {
92387
+ "epoch": 122.25,
92388
+ "learning_rate": 9.770468497576737e-06,
92389
+ "loss": 0.4715,
92390
+ "step": 15214
92391
+ },
92392
+ {
92393
+ "epoch": 122.26,
92394
+ "learning_rate": 9.770452342487885e-06,
92395
+ "loss": 0.7973,
92396
+ "step": 15215
92397
+ },
92398
+ {
92399
+ "epoch": 122.27,
92400
+ "learning_rate": 9.770436187399032e-06,
92401
+ "loss": 0.4523,
92402
+ "step": 15216
92403
+ },
92404
+ {
92405
+ "epoch": 122.27,
92406
+ "learning_rate": 9.770420032310179e-06,
92407
+ "loss": 0.4915,
92408
+ "step": 15217
92409
+ },
92410
+ {
92411
+ "epoch": 122.28,
92412
+ "learning_rate": 9.770403877221326e-06,
92413
+ "loss": 0.4732,
92414
+ "step": 15218
92415
+ },
92416
+ {
92417
+ "epoch": 122.29,
92418
+ "learning_rate": 9.770387722132472e-06,
92419
+ "loss": 0.7222,
92420
+ "step": 15219
92421
+ },
92422
+ {
92423
+ "epoch": 122.3,
92424
+ "learning_rate": 9.77037156704362e-06,
92425
+ "loss": 0.4522,
92426
+ "step": 15220
92427
+ },
92428
+ {
92429
+ "epoch": 122.31,
92430
+ "learning_rate": 9.770355411954766e-06,
92431
+ "loss": 0.4966,
92432
+ "step": 15221
92433
+ },
92434
+ {
92435
+ "epoch": 122.31,
92436
+ "learning_rate": 9.770339256865915e-06,
92437
+ "loss": 0.4216,
92438
+ "step": 15222
92439
+ },
92440
+ {
92441
+ "epoch": 122.32,
92442
+ "learning_rate": 9.770323101777061e-06,
92443
+ "loss": 1.2825,
92444
+ "step": 15223
92445
+ },
92446
+ {
92447
+ "epoch": 122.33,
92448
+ "learning_rate": 9.770306946688207e-06,
92449
+ "loss": 0.6619,
92450
+ "step": 15224
92451
+ },
92452
+ {
92453
+ "epoch": 122.34,
92454
+ "learning_rate": 9.770290791599355e-06,
92455
+ "loss": 0.6467,
92456
+ "step": 15225
92457
+ },
92458
+ {
92459
+ "epoch": 122.35,
92460
+ "learning_rate": 9.770274636510502e-06,
92461
+ "loss": 0.7571,
92462
+ "step": 15226
92463
+ },
92464
+ {
92465
+ "epoch": 122.35,
92466
+ "learning_rate": 9.770258481421649e-06,
92467
+ "loss": 0.8478,
92468
+ "step": 15227
92469
+ },
92470
+ {
92471
+ "epoch": 122.36,
92472
+ "learning_rate": 9.770242326332796e-06,
92473
+ "loss": 1.0236,
92474
+ "step": 15228
92475
+ },
92476
+ {
92477
+ "epoch": 122.37,
92478
+ "learning_rate": 9.770226171243942e-06,
92479
+ "loss": 0.9915,
92480
+ "step": 15229
92481
+ },
92482
+ {
92483
+ "epoch": 122.38,
92484
+ "learning_rate": 9.77021001615509e-06,
92485
+ "loss": 1.3893,
92486
+ "step": 15230
92487
+ },
92488
+ {
92489
+ "epoch": 122.39,
92490
+ "learning_rate": 9.770193861066236e-06,
92491
+ "loss": 2.1686,
92492
+ "step": 15231
92493
+ },
92494
+ {
92495
+ "epoch": 122.4,
92496
+ "learning_rate": 9.770177705977385e-06,
92497
+ "loss": 1.9717,
92498
+ "step": 15232
92499
+ },
92500
+ {
92501
+ "epoch": 122.4,
92502
+ "learning_rate": 9.770161550888531e-06,
92503
+ "loss": 2.5452,
92504
+ "step": 15233
92505
+ },
92506
+ {
92507
+ "epoch": 122.41,
92508
+ "learning_rate": 9.770145395799676e-06,
92509
+ "loss": 0.5942,
92510
+ "step": 15234
92511
+ },
92512
+ {
92513
+ "epoch": 122.42,
92514
+ "learning_rate": 9.770129240710825e-06,
92515
+ "loss": 0.5551,
92516
+ "step": 15235
92517
+ },
92518
+ {
92519
+ "epoch": 122.43,
92520
+ "learning_rate": 9.770113085621972e-06,
92521
+ "loss": 0.5168,
92522
+ "step": 15236
92523
+ },
92524
+ {
92525
+ "epoch": 122.44,
92526
+ "learning_rate": 9.770096930533119e-06,
92527
+ "loss": 0.5074,
92528
+ "step": 15237
92529
+ },
92530
+ {
92531
+ "epoch": 122.44,
92532
+ "learning_rate": 9.770080775444265e-06,
92533
+ "loss": 0.7036,
92534
+ "step": 15238
92535
+ },
92536
+ {
92537
+ "epoch": 122.45,
92538
+ "learning_rate": 9.770064620355412e-06,
92539
+ "loss": 0.5425,
92540
+ "step": 15239
92541
+ },
92542
+ {
92543
+ "epoch": 122.46,
92544
+ "learning_rate": 9.770048465266559e-06,
92545
+ "loss": 0.5473,
92546
+ "step": 15240
92547
+ },
92548
+ {
92549
+ "epoch": 122.47,
92550
+ "learning_rate": 9.770032310177708e-06,
92551
+ "loss": 0.4884,
92552
+ "step": 15241
92553
+ },
92554
+ {
92555
+ "epoch": 122.48,
92556
+ "learning_rate": 9.770016155088854e-06,
92557
+ "loss": 0.7163,
92558
+ "step": 15242
92559
+ },
92560
+ {
92561
+ "epoch": 122.48,
92562
+ "learning_rate": 9.770000000000001e-06,
92563
+ "loss": 0.5311,
92564
+ "step": 15243
92565
+ },
92566
+ {
92567
+ "epoch": 122.49,
92568
+ "learning_rate": 9.769983844911146e-06,
92569
+ "loss": 0.55,
92570
+ "step": 15244
92571
+ },
92572
+ {
92573
+ "epoch": 122.5,
92574
+ "learning_rate": 9.769967689822295e-06,
92575
+ "loss": 0.4682,
92576
+ "step": 15245
92577
+ },
92578
+ {
92579
+ "epoch": 122.51,
92580
+ "learning_rate": 9.769951534733442e-06,
92581
+ "loss": 0.4799,
92582
+ "step": 15246
92583
+ },
92584
+ {
92585
+ "epoch": 122.52,
92586
+ "learning_rate": 9.769935379644589e-06,
92587
+ "loss": 0.5239,
92588
+ "step": 15247
92589
+ },
92590
+ {
92591
+ "epoch": 122.52,
92592
+ "learning_rate": 9.769919224555735e-06,
92593
+ "loss": 0.6696,
92594
+ "step": 15248
92595
+ },
92596
+ {
92597
+ "epoch": 122.53,
92598
+ "learning_rate": 9.769903069466882e-06,
92599
+ "loss": 0.5761,
92600
+ "step": 15249
92601
+ },
92602
+ {
92603
+ "epoch": 122.54,
92604
+ "learning_rate": 9.769886914378029e-06,
92605
+ "loss": 0.647,
92606
+ "step": 15250
92607
+ },
92608
+ {
92609
+ "epoch": 122.55,
92610
+ "learning_rate": 9.769870759289178e-06,
92611
+ "loss": 0.7002,
92612
+ "step": 15251
92613
+ },
92614
+ {
92615
+ "epoch": 122.56,
92616
+ "learning_rate": 9.769854604200324e-06,
92617
+ "loss": 0.9307,
92618
+ "step": 15252
92619
+ },
92620
+ {
92621
+ "epoch": 122.56,
92622
+ "learning_rate": 9.769838449111471e-06,
92623
+ "loss": 0.7881,
92624
+ "step": 15253
92625
+ },
92626
+ {
92627
+ "epoch": 122.57,
92628
+ "learning_rate": 9.769822294022618e-06,
92629
+ "loss": 1.0559,
92630
+ "step": 15254
92631
+ },
92632
+ {
92633
+ "epoch": 122.58,
92634
+ "learning_rate": 9.769806138933765e-06,
92635
+ "loss": 1.2936,
92636
+ "step": 15255
92637
+ },
92638
+ {
92639
+ "epoch": 122.59,
92640
+ "learning_rate": 9.769789983844912e-06,
92641
+ "loss": 1.3611,
92642
+ "step": 15256
92643
+ },
92644
+ {
92645
+ "epoch": 122.6,
92646
+ "learning_rate": 9.76977382875606e-06,
92647
+ "loss": 1.4481,
92648
+ "step": 15257
92649
+ },
92650
+ {
92651
+ "epoch": 122.6,
92652
+ "learning_rate": 9.769757673667205e-06,
92653
+ "loss": 2.4536,
92654
+ "step": 15258
92655
+ },
92656
+ {
92657
+ "epoch": 122.61,
92658
+ "learning_rate": 9.769741518578352e-06,
92659
+ "loss": 0.7613,
92660
+ "step": 15259
92661
+ },
92662
+ {
92663
+ "epoch": 122.62,
92664
+ "learning_rate": 9.769725363489499e-06,
92665
+ "loss": 0.5705,
92666
+ "step": 15260
92667
+ },
92668
+ {
92669
+ "epoch": 122.63,
92670
+ "learning_rate": 9.769709208400648e-06,
92671
+ "loss": 0.6696,
92672
+ "step": 15261
92673
+ },
92674
+ {
92675
+ "epoch": 122.64,
92676
+ "learning_rate": 9.769693053311794e-06,
92677
+ "loss": 0.64,
92678
+ "step": 15262
92679
+ },
92680
+ {
92681
+ "epoch": 122.65,
92682
+ "learning_rate": 9.769676898222941e-06,
92683
+ "loss": 0.6128,
92684
+ "step": 15263
92685
+ },
92686
+ {
92687
+ "epoch": 122.65,
92688
+ "learning_rate": 9.769660743134088e-06,
92689
+ "loss": 0.4665,
92690
+ "step": 15264
92691
+ },
92692
+ {
92693
+ "epoch": 122.66,
92694
+ "learning_rate": 9.769644588045235e-06,
92695
+ "loss": 0.4298,
92696
+ "step": 15265
92697
+ },
92698
+ {
92699
+ "epoch": 122.67,
92700
+ "learning_rate": 9.769628432956382e-06,
92701
+ "loss": 0.5476,
92702
+ "step": 15266
92703
+ },
92704
+ {
92705
+ "epoch": 122.68,
92706
+ "learning_rate": 9.76961227786753e-06,
92707
+ "loss": 0.4967,
92708
+ "step": 15267
92709
+ },
92710
+ {
92711
+ "epoch": 122.69,
92712
+ "learning_rate": 9.769596122778675e-06,
92713
+ "loss": 0.4601,
92714
+ "step": 15268
92715
+ },
92716
+ {
92717
+ "epoch": 122.69,
92718
+ "learning_rate": 9.769579967689822e-06,
92719
+ "loss": 0.5816,
92720
+ "step": 15269
92721
+ },
92722
+ {
92723
+ "epoch": 122.7,
92724
+ "learning_rate": 9.76956381260097e-06,
92725
+ "loss": 0.5019,
92726
+ "step": 15270
92727
+ },
92728
+ {
92729
+ "epoch": 122.71,
92730
+ "learning_rate": 9.769547657512118e-06,
92731
+ "loss": 0.4007,
92732
+ "step": 15271
92733
+ },
92734
+ {
92735
+ "epoch": 122.72,
92736
+ "learning_rate": 9.769531502423264e-06,
92737
+ "loss": 0.5665,
92738
+ "step": 15272
92739
+ },
92740
+ {
92741
+ "epoch": 122.73,
92742
+ "learning_rate": 9.769515347334411e-06,
92743
+ "loss": 0.6853,
92744
+ "step": 15273
92745
+ },
92746
+ {
92747
+ "epoch": 122.73,
92748
+ "learning_rate": 9.769499192245558e-06,
92749
+ "loss": 0.4977,
92750
+ "step": 15274
92751
+ },
92752
+ {
92753
+ "epoch": 122.74,
92754
+ "learning_rate": 9.769483037156705e-06,
92755
+ "loss": 0.887,
92756
+ "step": 15275
92757
+ },
92758
+ {
92759
+ "epoch": 122.75,
92760
+ "learning_rate": 9.769466882067852e-06,
92761
+ "loss": 0.814,
92762
+ "step": 15276
92763
+ },
92764
+ {
92765
+ "epoch": 122.76,
92766
+ "learning_rate": 9.769450726979e-06,
92767
+ "loss": 0.5945,
92768
+ "step": 15277
92769
+ },
92770
+ {
92771
+ "epoch": 122.77,
92772
+ "learning_rate": 9.769434571890145e-06,
92773
+ "loss": 0.8064,
92774
+ "step": 15278
92775
+ },
92776
+ {
92777
+ "epoch": 122.77,
92778
+ "learning_rate": 9.769418416801292e-06,
92779
+ "loss": 0.9811,
92780
+ "step": 15279
92781
+ },
92782
+ {
92783
+ "epoch": 122.78,
92784
+ "learning_rate": 9.76940226171244e-06,
92785
+ "loss": 1.2227,
92786
+ "step": 15280
92787
+ },
92788
+ {
92789
+ "epoch": 122.79,
92790
+ "learning_rate": 9.769386106623588e-06,
92791
+ "loss": 1.4334,
92792
+ "step": 15281
92793
+ },
92794
+ {
92795
+ "epoch": 122.8,
92796
+ "learning_rate": 9.769369951534734e-06,
92797
+ "loss": 1.9314,
92798
+ "step": 15282
92799
+ },
92800
+ {
92801
+ "epoch": 122.81,
92802
+ "learning_rate": 9.769353796445881e-06,
92803
+ "loss": 2.3807,
92804
+ "step": 15283
92805
+ },
92806
+ {
92807
+ "epoch": 122.81,
92808
+ "learning_rate": 9.769337641357028e-06,
92809
+ "loss": 0.687,
92810
+ "step": 15284
92811
+ },
92812
+ {
92813
+ "epoch": 122.82,
92814
+ "learning_rate": 9.769321486268175e-06,
92815
+ "loss": 0.6531,
92816
+ "step": 15285
92817
+ },
92818
+ {
92819
+ "epoch": 122.83,
92820
+ "learning_rate": 9.769305331179322e-06,
92821
+ "loss": 0.4846,
92822
+ "step": 15286
92823
+ },
92824
+ {
92825
+ "epoch": 122.84,
92826
+ "learning_rate": 9.76928917609047e-06,
92827
+ "loss": 0.5867,
92828
+ "step": 15287
92829
+ },
92830
+ {
92831
+ "epoch": 122.85,
92832
+ "learning_rate": 9.769273021001617e-06,
92833
+ "loss": 0.6573,
92834
+ "step": 15288
92835
+ },
92836
+ {
92837
+ "epoch": 122.85,
92838
+ "learning_rate": 9.769256865912762e-06,
92839
+ "loss": 0.5612,
92840
+ "step": 15289
92841
+ },
92842
+ {
92843
+ "epoch": 122.86,
92844
+ "learning_rate": 9.76924071082391e-06,
92845
+ "loss": 1.1234,
92846
+ "step": 15290
92847
+ },
92848
+ {
92849
+ "epoch": 122.87,
92850
+ "learning_rate": 9.769224555735058e-06,
92851
+ "loss": 0.4894,
92852
+ "step": 15291
92853
+ },
92854
+ {
92855
+ "epoch": 122.88,
92856
+ "learning_rate": 9.769208400646204e-06,
92857
+ "loss": 0.4911,
92858
+ "step": 15292
92859
+ },
92860
+ {
92861
+ "epoch": 122.89,
92862
+ "learning_rate": 9.769192245557351e-06,
92863
+ "loss": 0.6806,
92864
+ "step": 15293
92865
+ },
92866
+ {
92867
+ "epoch": 122.9,
92868
+ "learning_rate": 9.769176090468498e-06,
92869
+ "loss": 0.5802,
92870
+ "step": 15294
92871
+ },
92872
+ {
92873
+ "epoch": 122.9,
92874
+ "learning_rate": 9.769159935379645e-06,
92875
+ "loss": 0.5569,
92876
+ "step": 15295
92877
+ },
92878
+ {
92879
+ "epoch": 122.91,
92880
+ "learning_rate": 9.769143780290793e-06,
92881
+ "loss": 0.6317,
92882
+ "step": 15296
92883
+ },
92884
+ {
92885
+ "epoch": 122.92,
92886
+ "learning_rate": 9.76912762520194e-06,
92887
+ "loss": 0.5695,
92888
+ "step": 15297
92889
+ },
92890
+ {
92891
+ "epoch": 122.93,
92892
+ "learning_rate": 9.769111470113087e-06,
92893
+ "loss": 0.6493,
92894
+ "step": 15298
92895
+ },
92896
+ {
92897
+ "epoch": 122.94,
92898
+ "learning_rate": 9.769095315024232e-06,
92899
+ "loss": 0.6368,
92900
+ "step": 15299
92901
+ },
92902
+ {
92903
+ "epoch": 122.94,
92904
+ "learning_rate": 9.76907915993538e-06,
92905
+ "loss": 0.6335,
92906
+ "step": 15300
92907
+ },
92908
+ {
92909
+ "epoch": 122.95,
92910
+ "learning_rate": 9.769063004846528e-06,
92911
+ "loss": 0.7491,
92912
+ "step": 15301
92913
+ },
92914
+ {
92915
+ "epoch": 122.96,
92916
+ "learning_rate": 9.769046849757674e-06,
92917
+ "loss": 0.9096,
92918
+ "step": 15302
92919
+ },
92920
+ {
92921
+ "epoch": 122.97,
92922
+ "learning_rate": 9.769030694668821e-06,
92923
+ "loss": 1.2254,
92924
+ "step": 15303
92925
+ },
92926
+ {
92927
+ "epoch": 122.98,
92928
+ "learning_rate": 9.769014539579968e-06,
92929
+ "loss": 1.2207,
92930
+ "step": 15304
92931
+ },
92932
+ {
92933
+ "epoch": 122.98,
92934
+ "learning_rate": 9.768998384491115e-06,
92935
+ "loss": 1.2205,
92936
+ "step": 15305
92937
+ },
92938
+ {
92939
+ "epoch": 122.99,
92940
+ "learning_rate": 9.768982229402263e-06,
92941
+ "loss": 1.6907,
92942
+ "step": 15306
92943
+ },
92944
+ {
92945
+ "epoch": 123.0,
92946
+ "learning_rate": 9.76896607431341e-06,
92947
+ "loss": 2.108,
92948
+ "step": 15307
92949
+ },
92950
+ {
92951
+ "epoch": 123.0,
92952
+ "eval_loss": 0.6208506226539612,
92953
+ "eval_runtime": 39.7896,
92954
+ "eval_samples_per_second": 21.036,
92955
+ "eval_steps_per_second": 0.679,
92956
+ "eval_wer": 0.27671094426797577,
92957
+ "step": 15307
92958
+ },
92959
+ {
92960
+ "epoch": 123.01,
92961
+ "learning_rate": 9.768949919224557e-06,
92962
+ "loss": 0.6673,
92963
+ "step": 15308
92964
+ },
92965
+ {
92966
+ "epoch": 123.02,
92967
+ "learning_rate": 9.768933764135704e-06,
92968
+ "loss": 0.4808,
92969
+ "step": 15309
92970
+ },
92971
+ {
92972
+ "epoch": 123.02,
92973
+ "learning_rate": 9.76891760904685e-06,
92974
+ "loss": 0.4821,
92975
+ "step": 15310
92976
+ },
92977
+ {
92978
+ "epoch": 123.03,
92979
+ "learning_rate": 9.768901453957997e-06,
92980
+ "loss": 0.5539,
92981
+ "step": 15311
92982
+ },
92983
+ {
92984
+ "epoch": 123.04,
92985
+ "learning_rate": 9.768885298869144e-06,
92986
+ "loss": 0.5555,
92987
+ "step": 15312
92988
+ },
92989
+ {
92990
+ "epoch": 123.05,
92991
+ "learning_rate": 9.768869143780291e-06,
92992
+ "loss": 0.4294,
92993
+ "step": 15313
92994
+ },
92995
+ {
92996
+ "epoch": 123.06,
92997
+ "learning_rate": 9.768852988691438e-06,
92998
+ "loss": 0.5274,
92999
+ "step": 15314
93000
+ },
93001
+ {
93002
+ "epoch": 123.06,
93003
+ "learning_rate": 9.768836833602585e-06,
93004
+ "loss": 0.4149,
93005
+ "step": 15315
93006
+ },
93007
+ {
93008
+ "epoch": 123.07,
93009
+ "learning_rate": 9.768820678513733e-06,
93010
+ "loss": 0.4846,
93011
+ "step": 15316
93012
+ },
93013
+ {
93014
+ "epoch": 123.08,
93015
+ "learning_rate": 9.76880452342488e-06,
93016
+ "loss": 0.5718,
93017
+ "step": 15317
93018
+ },
93019
+ {
93020
+ "epoch": 123.09,
93021
+ "learning_rate": 9.768788368336027e-06,
93022
+ "loss": 0.7085,
93023
+ "step": 15318
93024
+ },
93025
+ {
93026
+ "epoch": 123.1,
93027
+ "learning_rate": 9.768772213247174e-06,
93028
+ "loss": 0.5464,
93029
+ "step": 15319
93030
+ },
93031
+ {
93032
+ "epoch": 123.1,
93033
+ "learning_rate": 9.76875605815832e-06,
93034
+ "loss": 0.8453,
93035
+ "step": 15320
93036
+ },
93037
+ {
93038
+ "epoch": 123.11,
93039
+ "learning_rate": 9.768739903069467e-06,
93040
+ "loss": 0.4346,
93041
+ "step": 15321
93042
+ },
93043
+ {
93044
+ "epoch": 123.12,
93045
+ "learning_rate": 9.768723747980616e-06,
93046
+ "loss": 0.605,
93047
+ "step": 15322
93048
+ },
93049
+ {
93050
+ "epoch": 123.13,
93051
+ "learning_rate": 9.768707592891761e-06,
93052
+ "loss": 0.6585,
93053
+ "step": 15323
93054
+ },
93055
+ {
93056
+ "epoch": 123.14,
93057
+ "learning_rate": 9.768691437802908e-06,
93058
+ "loss": 1.1555,
93059
+ "step": 15324
93060
+ },
93061
+ {
93062
+ "epoch": 123.15,
93063
+ "learning_rate": 9.768675282714055e-06,
93064
+ "loss": 0.812,
93065
+ "step": 15325
93066
+ },
93067
+ {
93068
+ "epoch": 123.15,
93069
+ "learning_rate": 9.768659127625203e-06,
93070
+ "loss": 0.6161,
93071
+ "step": 15326
93072
+ },
93073
+ {
93074
+ "epoch": 123.16,
93075
+ "learning_rate": 9.76864297253635e-06,
93076
+ "loss": 1.0263,
93077
+ "step": 15327
93078
+ },
93079
+ {
93080
+ "epoch": 123.17,
93081
+ "learning_rate": 9.768626817447497e-06,
93082
+ "loss": 1.0453,
93083
+ "step": 15328
93084
+ },
93085
+ {
93086
+ "epoch": 123.18,
93087
+ "learning_rate": 9.768610662358644e-06,
93088
+ "loss": 1.2194,
93089
+ "step": 15329
93090
+ },
93091
+ {
93092
+ "epoch": 123.19,
93093
+ "learning_rate": 9.76859450726979e-06,
93094
+ "loss": 1.43,
93095
+ "step": 15330
93096
+ },
93097
+ {
93098
+ "epoch": 123.19,
93099
+ "learning_rate": 9.768578352180937e-06,
93100
+ "loss": 2.2039,
93101
+ "step": 15331
93102
+ },
93103
+ {
93104
+ "epoch": 123.2,
93105
+ "learning_rate": 9.768562197092086e-06,
93106
+ "loss": 2.7584,
93107
+ "step": 15332
93108
+ },
93109
+ {
93110
+ "epoch": 123.21,
93111
+ "learning_rate": 9.768546042003231e-06,
93112
+ "loss": 0.7303,
93113
+ "step": 15333
93114
+ },
93115
+ {
93116
+ "epoch": 123.22,
93117
+ "learning_rate": 9.768529886914378e-06,
93118
+ "loss": 0.6692,
93119
+ "step": 15334
93120
+ },
93121
+ {
93122
+ "epoch": 123.23,
93123
+ "learning_rate": 9.768513731825526e-06,
93124
+ "loss": 0.5593,
93125
+ "step": 15335
93126
+ },
93127
+ {
93128
+ "epoch": 123.23,
93129
+ "learning_rate": 9.768497576736673e-06,
93130
+ "loss": 0.5257,
93131
+ "step": 15336
93132
+ },
93133
+ {
93134
+ "epoch": 123.24,
93135
+ "learning_rate": 9.76848142164782e-06,
93136
+ "loss": 0.8428,
93137
+ "step": 15337
93138
+ },
93139
+ {
93140
+ "epoch": 123.25,
93141
+ "learning_rate": 9.768465266558967e-06,
93142
+ "loss": 0.4554,
93143
+ "step": 15338
93144
+ },
93145
+ {
93146
+ "epoch": 123.26,
93147
+ "learning_rate": 9.768449111470114e-06,
93148
+ "loss": 0.5604,
93149
+ "step": 15339
93150
+ },
93151
+ {
93152
+ "epoch": 123.27,
93153
+ "learning_rate": 9.76843295638126e-06,
93154
+ "loss": 0.5804,
93155
+ "step": 15340
93156
+ },
93157
+ {
93158
+ "epoch": 123.27,
93159
+ "learning_rate": 9.768416801292407e-06,
93160
+ "loss": 0.543,
93161
+ "step": 15341
93162
+ },
93163
+ {
93164
+ "epoch": 123.28,
93165
+ "learning_rate": 9.768400646203556e-06,
93166
+ "loss": 0.4608,
93167
+ "step": 15342
93168
+ },
93169
+ {
93170
+ "epoch": 123.29,
93171
+ "learning_rate": 9.768384491114701e-06,
93172
+ "loss": 0.5164,
93173
+ "step": 15343
93174
+ },
93175
+ {
93176
+ "epoch": 123.3,
93177
+ "learning_rate": 9.768368336025848e-06,
93178
+ "loss": 0.5764,
93179
+ "step": 15344
93180
+ },
93181
+ {
93182
+ "epoch": 123.31,
93183
+ "learning_rate": 9.768352180936996e-06,
93184
+ "loss": 0.5852,
93185
+ "step": 15345
93186
+ },
93187
+ {
93188
+ "epoch": 123.31,
93189
+ "learning_rate": 9.768336025848143e-06,
93190
+ "loss": 0.5773,
93191
+ "step": 15346
93192
+ },
93193
+ {
93194
+ "epoch": 123.32,
93195
+ "learning_rate": 9.76831987075929e-06,
93196
+ "loss": 0.7063,
93197
+ "step": 15347
93198
+ },
93199
+ {
93200
+ "epoch": 123.33,
93201
+ "learning_rate": 9.768303715670437e-06,
93202
+ "loss": 0.6608,
93203
+ "step": 15348
93204
+ },
93205
+ {
93206
+ "epoch": 123.34,
93207
+ "learning_rate": 9.768287560581584e-06,
93208
+ "loss": 0.6202,
93209
+ "step": 15349
93210
+ },
93211
+ {
93212
+ "epoch": 123.35,
93213
+ "learning_rate": 9.76827140549273e-06,
93214
+ "loss": 0.8583,
93215
+ "step": 15350
93216
+ },
93217
+ {
93218
+ "epoch": 123.35,
93219
+ "learning_rate": 9.768255250403879e-06,
93220
+ "loss": 0.702,
93221
+ "step": 15351
93222
+ },
93223
+ {
93224
+ "epoch": 123.36,
93225
+ "learning_rate": 9.768239095315026e-06,
93226
+ "loss": 1.0671,
93227
+ "step": 15352
93228
+ },
93229
+ {
93230
+ "epoch": 123.37,
93231
+ "learning_rate": 9.768222940226173e-06,
93232
+ "loss": 1.4001,
93233
+ "step": 15353
93234
+ },
93235
+ {
93236
+ "epoch": 123.38,
93237
+ "learning_rate": 9.768206785137318e-06,
93238
+ "loss": 1.1794,
93239
+ "step": 15354
93240
+ },
93241
+ {
93242
+ "epoch": 123.39,
93243
+ "learning_rate": 9.768190630048466e-06,
93244
+ "loss": 1.5339,
93245
+ "step": 15355
93246
+ },
93247
+ {
93248
+ "epoch": 123.4,
93249
+ "learning_rate": 9.768174474959613e-06,
93250
+ "loss": 1.7817,
93251
+ "step": 15356
93252
+ },
93253
+ {
93254
+ "epoch": 123.4,
93255
+ "learning_rate": 9.76815831987076e-06,
93256
+ "loss": 2.8118,
93257
+ "step": 15357
93258
+ },
93259
+ {
93260
+ "epoch": 123.41,
93261
+ "learning_rate": 9.768142164781907e-06,
93262
+ "loss": 0.8838,
93263
+ "step": 15358
93264
+ },
93265
+ {
93266
+ "epoch": 123.42,
93267
+ "learning_rate": 9.768126009693054e-06,
93268
+ "loss": 0.562,
93269
+ "step": 15359
93270
+ },
93271
+ {
93272
+ "epoch": 123.43,
93273
+ "learning_rate": 9.7681098546042e-06,
93274
+ "loss": 0.6152,
93275
+ "step": 15360
93276
+ },
93277
+ {
93278
+ "epoch": 123.44,
93279
+ "learning_rate": 9.768093699515349e-06,
93280
+ "loss": 0.5278,
93281
+ "step": 15361
93282
+ },
93283
+ {
93284
+ "epoch": 123.44,
93285
+ "learning_rate": 9.768077544426496e-06,
93286
+ "loss": 0.539,
93287
+ "step": 15362
93288
+ },
93289
+ {
93290
+ "epoch": 123.45,
93291
+ "learning_rate": 9.768061389337643e-06,
93292
+ "loss": 0.4325,
93293
+ "step": 15363
93294
+ },
93295
+ {
93296
+ "epoch": 123.46,
93297
+ "learning_rate": 9.76804523424879e-06,
93298
+ "loss": 0.5053,
93299
+ "step": 15364
93300
+ },
93301
+ {
93302
+ "epoch": 123.47,
93303
+ "learning_rate": 9.768029079159936e-06,
93304
+ "loss": 0.4931,
93305
+ "step": 15365
93306
+ },
93307
+ {
93308
+ "epoch": 123.48,
93309
+ "learning_rate": 9.768012924071083e-06,
93310
+ "loss": 0.5907,
93311
+ "step": 15366
93312
+ },
93313
+ {
93314
+ "epoch": 123.48,
93315
+ "learning_rate": 9.76799676898223e-06,
93316
+ "loss": 0.528,
93317
+ "step": 15367
93318
+ },
93319
+ {
93320
+ "epoch": 123.49,
93321
+ "learning_rate": 9.767980613893377e-06,
93322
+ "loss": 0.5645,
93323
+ "step": 15368
93324
+ },
93325
+ {
93326
+ "epoch": 123.5,
93327
+ "learning_rate": 9.767964458804524e-06,
93328
+ "loss": 0.5449,
93329
+ "step": 15369
93330
+ },
93331
+ {
93332
+ "epoch": 123.51,
93333
+ "learning_rate": 9.76794830371567e-06,
93334
+ "loss": 0.7076,
93335
+ "step": 15370
93336
+ },
93337
+ {
93338
+ "epoch": 123.52,
93339
+ "learning_rate": 9.767932148626819e-06,
93340
+ "loss": 0.58,
93341
+ "step": 15371
93342
+ },
93343
+ {
93344
+ "epoch": 123.52,
93345
+ "learning_rate": 9.767915993537966e-06,
93346
+ "loss": 0.7,
93347
+ "step": 15372
93348
+ },
93349
+ {
93350
+ "epoch": 123.53,
93351
+ "learning_rate": 9.767899838449113e-06,
93352
+ "loss": 0.6149,
93353
+ "step": 15373
93354
+ },
93355
+ {
93356
+ "epoch": 123.54,
93357
+ "learning_rate": 9.76788368336026e-06,
93358
+ "loss": 0.5157,
93359
+ "step": 15374
93360
+ },
93361
+ {
93362
+ "epoch": 123.55,
93363
+ "learning_rate": 9.767867528271406e-06,
93364
+ "loss": 0.5652,
93365
+ "step": 15375
93366
+ },
93367
+ {
93368
+ "epoch": 123.56,
93369
+ "learning_rate": 9.767851373182553e-06,
93370
+ "loss": 0.8276,
93371
+ "step": 15376
93372
+ },
93373
+ {
93374
+ "epoch": 123.56,
93375
+ "learning_rate": 9.7678352180937e-06,
93376
+ "loss": 0.9,
93377
+ "step": 15377
93378
+ },
93379
+ {
93380
+ "epoch": 123.57,
93381
+ "learning_rate": 9.767819063004847e-06,
93382
+ "loss": 0.9305,
93383
+ "step": 15378
93384
+ },
93385
+ {
93386
+ "epoch": 123.58,
93387
+ "learning_rate": 9.767802907915994e-06,
93388
+ "loss": 1.0592,
93389
+ "step": 15379
93390
+ },
93391
+ {
93392
+ "epoch": 123.59,
93393
+ "learning_rate": 9.76778675282714e-06,
93394
+ "loss": 1.6534,
93395
+ "step": 15380
93396
+ },
93397
+ {
93398
+ "epoch": 123.6,
93399
+ "learning_rate": 9.767770597738289e-06,
93400
+ "loss": 2.4195,
93401
+ "step": 15381
93402
+ },
93403
+ {
93404
+ "epoch": 123.6,
93405
+ "learning_rate": 9.767754442649436e-06,
93406
+ "loss": 2.2875,
93407
+ "step": 15382
93408
+ },
93409
+ {
93410
+ "epoch": 123.61,
93411
+ "learning_rate": 9.767738287560583e-06,
93412
+ "loss": 0.6068,
93413
+ "step": 15383
93414
+ },
93415
+ {
93416
+ "epoch": 123.62,
93417
+ "learning_rate": 9.76772213247173e-06,
93418
+ "loss": 0.5069,
93419
+ "step": 15384
93420
+ },
93421
+ {
93422
+ "epoch": 123.63,
93423
+ "learning_rate": 9.767705977382876e-06,
93424
+ "loss": 0.6553,
93425
+ "step": 15385
93426
+ },
93427
+ {
93428
+ "epoch": 123.64,
93429
+ "learning_rate": 9.767689822294023e-06,
93430
+ "loss": 0.4451,
93431
+ "step": 15386
93432
+ },
93433
+ {
93434
+ "epoch": 123.65,
93435
+ "learning_rate": 9.767673667205172e-06,
93436
+ "loss": 0.4952,
93437
+ "step": 15387
93438
+ },
93439
+ {
93440
+ "epoch": 123.65,
93441
+ "learning_rate": 9.767657512116317e-06,
93442
+ "loss": 0.4616,
93443
+ "step": 15388
93444
+ },
93445
+ {
93446
+ "epoch": 123.66,
93447
+ "learning_rate": 9.767641357027464e-06,
93448
+ "loss": 0.4906,
93449
+ "step": 15389
93450
+ },
93451
+ {
93452
+ "epoch": 123.67,
93453
+ "learning_rate": 9.767625201938612e-06,
93454
+ "loss": 0.6653,
93455
+ "step": 15390
93456
+ },
93457
+ {
93458
+ "epoch": 123.68,
93459
+ "learning_rate": 9.767609046849759e-06,
93460
+ "loss": 0.3601,
93461
+ "step": 15391
93462
+ },
93463
+ {
93464
+ "epoch": 123.69,
93465
+ "learning_rate": 9.767592891760906e-06,
93466
+ "loss": 0.6886,
93467
+ "step": 15392
93468
+ },
93469
+ {
93470
+ "epoch": 123.69,
93471
+ "learning_rate": 9.767576736672053e-06,
93472
+ "loss": 0.4421,
93473
+ "step": 15393
93474
+ },
93475
+ {
93476
+ "epoch": 123.7,
93477
+ "learning_rate": 9.7675605815832e-06,
93478
+ "loss": 0.4743,
93479
+ "step": 15394
93480
+ },
93481
+ {
93482
+ "epoch": 123.71,
93483
+ "learning_rate": 9.767544426494346e-06,
93484
+ "loss": 0.637,
93485
+ "step": 15395
93486
+ },
93487
+ {
93488
+ "epoch": 123.72,
93489
+ "learning_rate": 9.767528271405493e-06,
93490
+ "loss": 0.8284,
93491
+ "step": 15396
93492
+ },
93493
+ {
93494
+ "epoch": 123.73,
93495
+ "learning_rate": 9.767512116316642e-06,
93496
+ "loss": 0.6071,
93497
+ "step": 15397
93498
+ },
93499
+ {
93500
+ "epoch": 123.73,
93501
+ "learning_rate": 9.767495961227787e-06,
93502
+ "loss": 0.5572,
93503
+ "step": 15398
93504
+ },
93505
+ {
93506
+ "epoch": 123.74,
93507
+ "learning_rate": 9.767479806138934e-06,
93508
+ "loss": 0.6809,
93509
+ "step": 15399
93510
+ },
93511
+ {
93512
+ "epoch": 123.75,
93513
+ "learning_rate": 9.767463651050082e-06,
93514
+ "loss": 0.5209,
93515
+ "step": 15400
93516
+ },
93517
+ {
93518
+ "epoch": 123.76,
93519
+ "learning_rate": 9.767447495961229e-06,
93520
+ "loss": 0.8877,
93521
+ "step": 15401
93522
+ },
93523
+ {
93524
+ "epoch": 123.77,
93525
+ "learning_rate": 9.767431340872376e-06,
93526
+ "loss": 1.0002,
93527
+ "step": 15402
93528
+ },
93529
+ {
93530
+ "epoch": 123.77,
93531
+ "learning_rate": 9.767415185783523e-06,
93532
+ "loss": 1.1209,
93533
+ "step": 15403
93534
+ },
93535
+ {
93536
+ "epoch": 123.78,
93537
+ "learning_rate": 9.76739903069467e-06,
93538
+ "loss": 1.1321,
93539
+ "step": 15404
93540
+ },
93541
+ {
93542
+ "epoch": 123.79,
93543
+ "learning_rate": 9.767382875605816e-06,
93544
+ "loss": 1.3916,
93545
+ "step": 15405
93546
+ },
93547
+ {
93548
+ "epoch": 123.8,
93549
+ "learning_rate": 9.767366720516963e-06,
93550
+ "loss": 2.0445,
93551
+ "step": 15406
93552
+ },
93553
+ {
93554
+ "epoch": 123.81,
93555
+ "learning_rate": 9.767350565428112e-06,
93556
+ "loss": 2.3018,
93557
+ "step": 15407
93558
+ },
93559
+ {
93560
+ "epoch": 123.81,
93561
+ "learning_rate": 9.767334410339257e-06,
93562
+ "loss": 0.6054,
93563
+ "step": 15408
93564
+ },
93565
+ {
93566
+ "epoch": 123.82,
93567
+ "learning_rate": 9.767318255250404e-06,
93568
+ "loss": 0.8411,
93569
+ "step": 15409
93570
+ },
93571
+ {
93572
+ "epoch": 123.83,
93573
+ "learning_rate": 9.767302100161552e-06,
93574
+ "loss": 0.6948,
93575
+ "step": 15410
93576
+ },
93577
+ {
93578
+ "epoch": 123.84,
93579
+ "learning_rate": 9.767285945072699e-06,
93580
+ "loss": 0.4658,
93581
+ "step": 15411
93582
+ },
93583
+ {
93584
+ "epoch": 123.85,
93585
+ "learning_rate": 9.767269789983846e-06,
93586
+ "loss": 0.4538,
93587
+ "step": 15412
93588
+ },
93589
+ {
93590
+ "epoch": 123.85,
93591
+ "learning_rate": 9.767253634894993e-06,
93592
+ "loss": 0.7092,
93593
+ "step": 15413
93594
+ },
93595
+ {
93596
+ "epoch": 123.86,
93597
+ "learning_rate": 9.76723747980614e-06,
93598
+ "loss": 0.536,
93599
+ "step": 15414
93600
+ },
93601
+ {
93602
+ "epoch": 123.87,
93603
+ "learning_rate": 9.767221324717286e-06,
93604
+ "loss": 0.4595,
93605
+ "step": 15415
93606
+ },
93607
+ {
93608
+ "epoch": 123.88,
93609
+ "learning_rate": 9.767205169628435e-06,
93610
+ "loss": 0.5701,
93611
+ "step": 15416
93612
+ },
93613
+ {
93614
+ "epoch": 123.89,
93615
+ "learning_rate": 9.767189014539582e-06,
93616
+ "loss": 0.9473,
93617
+ "step": 15417
93618
+ },
93619
+ {
93620
+ "epoch": 123.9,
93621
+ "learning_rate": 9.767172859450727e-06,
93622
+ "loss": 0.4991,
93623
+ "step": 15418
93624
+ },
93625
+ {
93626
+ "epoch": 123.9,
93627
+ "learning_rate": 9.767156704361874e-06,
93628
+ "loss": 0.628,
93629
+ "step": 15419
93630
+ },
93631
+ {
93632
+ "epoch": 123.91,
93633
+ "learning_rate": 9.767140549273022e-06,
93634
+ "loss": 1.1212,
93635
+ "step": 15420
93636
+ },
93637
+ {
93638
+ "epoch": 123.92,
93639
+ "learning_rate": 9.767124394184169e-06,
93640
+ "loss": 0.8238,
93641
+ "step": 15421
93642
+ },
93643
+ {
93644
+ "epoch": 123.93,
93645
+ "learning_rate": 9.767108239095316e-06,
93646
+ "loss": 0.5751,
93647
+ "step": 15422
93648
+ },
93649
+ {
93650
+ "epoch": 123.94,
93651
+ "learning_rate": 9.767092084006463e-06,
93652
+ "loss": 0.6398,
93653
+ "step": 15423
93654
+ },
93655
+ {
93656
+ "epoch": 123.94,
93657
+ "learning_rate": 9.76707592891761e-06,
93658
+ "loss": 1.1882,
93659
+ "step": 15424
93660
+ },
93661
+ {
93662
+ "epoch": 123.95,
93663
+ "learning_rate": 9.767059773828756e-06,
93664
+ "loss": 0.6789,
93665
+ "step": 15425
93666
+ },
93667
+ {
93668
+ "epoch": 123.96,
93669
+ "learning_rate": 9.767043618739905e-06,
93670
+ "loss": 0.806,
93671
+ "step": 15426
93672
+ },
93673
+ {
93674
+ "epoch": 123.97,
93675
+ "learning_rate": 9.767027463651052e-06,
93676
+ "loss": 1.0979,
93677
+ "step": 15427
93678
+ },
93679
+ {
93680
+ "epoch": 123.98,
93681
+ "learning_rate": 9.767011308562198e-06,
93682
+ "loss": 1.2586,
93683
+ "step": 15428
93684
+ },
93685
+ {
93686
+ "epoch": 123.98,
93687
+ "learning_rate": 9.766995153473345e-06,
93688
+ "loss": 1.3842,
93689
+ "step": 15429
93690
+ },
93691
+ {
93692
+ "epoch": 123.99,
93693
+ "learning_rate": 9.766978998384492e-06,
93694
+ "loss": 1.8053,
93695
+ "step": 15430
93696
+ },
93697
+ {
93698
+ "epoch": 124.0,
93699
+ "learning_rate": 9.766962843295639e-06,
93700
+ "loss": 2.3601,
93701
+ "step": 15431
93702
+ },
93703
+ {
93704
+ "epoch": 124.0,
93705
+ "eval_loss": 0.6292822360992432,
93706
+ "eval_runtime": 39.9508,
93707
+ "eval_samples_per_second": 20.951,
93708
+ "eval_steps_per_second": 0.676,
93709
+ "eval_wer": 0.27080281277468504,
93710
+ "step": 15431
93711
+ },
93712
+ {
93713
+ "epoch": 123.01,
93714
+ "learning_rate": 9.766946688206786e-06,
93715
+ "loss": 0.6155,
93716
+ "step": 15432
93717
+ },
93718
+ {
93719
+ "epoch": 123.02,
93720
+ "learning_rate": 9.766930533117933e-06,
93721
+ "loss": 0.6373,
93722
+ "step": 15433
93723
+ },
93724
+ {
93725
+ "epoch": 123.02,
93726
+ "learning_rate": 9.76691437802908e-06,
93727
+ "loss": 0.6673,
93728
+ "step": 15434
93729
+ },
93730
+ {
93731
+ "epoch": 123.03,
93732
+ "learning_rate": 9.766898222940226e-06,
93733
+ "loss": 0.5828,
93734
+ "step": 15435
93735
+ },
93736
+ {
93737
+ "epoch": 123.04,
93738
+ "learning_rate": 9.766882067851375e-06,
93739
+ "loss": 0.4998,
93740
+ "step": 15436
93741
+ },
93742
+ {
93743
+ "epoch": 123.05,
93744
+ "learning_rate": 9.766865912762522e-06,
93745
+ "loss": 0.7833,
93746
+ "step": 15437
93747
+ },
93748
+ {
93749
+ "epoch": 123.06,
93750
+ "learning_rate": 9.766849757673668e-06,
93751
+ "loss": 0.6736,
93752
+ "step": 15438
93753
+ },
93754
+ {
93755
+ "epoch": 123.06,
93756
+ "learning_rate": 9.766833602584815e-06,
93757
+ "loss": 0.4026,
93758
+ "step": 15439
93759
+ },
93760
+ {
93761
+ "epoch": 123.07,
93762
+ "learning_rate": 9.766817447495962e-06,
93763
+ "loss": 0.4836,
93764
+ "step": 15440
93765
+ },
93766
+ {
93767
+ "epoch": 123.08,
93768
+ "learning_rate": 9.766801292407109e-06,
93769
+ "loss": 0.5984,
93770
+ "step": 15441
93771
+ },
93772
+ {
93773
+ "epoch": 123.09,
93774
+ "learning_rate": 9.766785137318256e-06,
93775
+ "loss": 0.6241,
93776
+ "step": 15442
93777
+ },
93778
+ {
93779
+ "epoch": 123.1,
93780
+ "learning_rate": 9.766768982229403e-06,
93781
+ "loss": 0.6752,
93782
+ "step": 15443
93783
+ },
93784
+ {
93785
+ "epoch": 123.1,
93786
+ "learning_rate": 9.76675282714055e-06,
93787
+ "loss": 0.5746,
93788
+ "step": 15444
93789
+ },
93790
+ {
93791
+ "epoch": 123.11,
93792
+ "learning_rate": 9.766736672051698e-06,
93793
+ "loss": 0.5676,
93794
+ "step": 15445
93795
+ },
93796
+ {
93797
+ "epoch": 123.12,
93798
+ "learning_rate": 9.766720516962845e-06,
93799
+ "loss": 0.8624,
93800
+ "step": 15446
93801
+ },
93802
+ {
93803
+ "epoch": 123.13,
93804
+ "learning_rate": 9.766704361873992e-06,
93805
+ "loss": 0.7062,
93806
+ "step": 15447
93807
+ },
93808
+ {
93809
+ "epoch": 123.14,
93810
+ "learning_rate": 9.766688206785138e-06,
93811
+ "loss": 0.5844,
93812
+ "step": 15448
93813
+ },
93814
+ {
93815
+ "epoch": 123.14,
93816
+ "learning_rate": 9.766672051696285e-06,
93817
+ "loss": 0.5917,
93818
+ "step": 15449
93819
+ },
93820
+ {
93821
+ "epoch": 123.15,
93822
+ "learning_rate": 9.766655896607432e-06,
93823
+ "loss": 1.0301,
93824
+ "step": 15450
93825
+ },
93826
+ {
93827
+ "epoch": 123.16,
93828
+ "learning_rate": 9.766639741518579e-06,
93829
+ "loss": 1.0731,
93830
+ "step": 15451
93831
+ },
93832
+ {
93833
+ "epoch": 123.17,
93834
+ "learning_rate": 9.766623586429727e-06,
93835
+ "loss": 1.0773,
93836
+ "step": 15452
93837
+ },
93838
+ {
93839
+ "epoch": 123.18,
93840
+ "learning_rate": 9.766607431340872e-06,
93841
+ "loss": 1.2137,
93842
+ "step": 15453
93843
+ },
93844
+ {
93845
+ "epoch": 123.18,
93846
+ "learning_rate": 9.76659127625202e-06,
93847
+ "loss": 1.4435,
93848
+ "step": 15454
93849
+ },
93850
+ {
93851
+ "epoch": 123.19,
93852
+ "learning_rate": 9.766575121163168e-06,
93853
+ "loss": 1.6229,
93854
+ "step": 15455
93855
+ },
93856
+ {
93857
+ "epoch": 123.2,
93858
+ "learning_rate": 9.766558966074315e-06,
93859
+ "loss": 2.4803,
93860
+ "step": 15456
93861
+ },
93862
+ {
93863
+ "epoch": 123.21,
93864
+ "learning_rate": 9.766542810985462e-06,
93865
+ "loss": 0.6243,
93866
+ "step": 15457
93867
+ },
93868
+ {
93869
+ "epoch": 123.22,
93870
+ "learning_rate": 9.766526655896608e-06,
93871
+ "loss": 1.202,
93872
+ "step": 15458
93873
+ },
93874
+ {
93875
+ "epoch": 123.22,
93876
+ "learning_rate": 9.766510500807755e-06,
93877
+ "loss": 0.6218,
93878
+ "step": 15459
93879
+ },
93880
+ {
93881
+ "epoch": 123.23,
93882
+ "learning_rate": 9.766494345718902e-06,
93883
+ "loss": 0.6766,
93884
+ "step": 15460
93885
+ },
93886
+ {
93887
+ "epoch": 123.24,
93888
+ "learning_rate": 9.766478190630049e-06,
93889
+ "loss": 0.6821,
93890
+ "step": 15461
93891
+ },
93892
+ {
93893
+ "epoch": 123.25,
93894
+ "learning_rate": 9.766462035541197e-06,
93895
+ "loss": 0.7872,
93896
+ "step": 15462
93897
+ },
93898
+ {
93899
+ "epoch": 123.26,
93900
+ "learning_rate": 9.766445880452342e-06,
93901
+ "loss": 0.4612,
93902
+ "step": 15463
93903
+ },
93904
+ {
93905
+ "epoch": 123.26,
93906
+ "learning_rate": 9.76642972536349e-06,
93907
+ "loss": 0.4166,
93908
+ "step": 15464
93909
+ },
93910
+ {
93911
+ "epoch": 123.27,
93912
+ "learning_rate": 9.766413570274638e-06,
93913
+ "loss": 0.4442,
93914
+ "step": 15465
93915
+ },
93916
+ {
93917
+ "epoch": 123.28,
93918
+ "learning_rate": 9.766397415185785e-06,
93919
+ "loss": 0.4603,
93920
+ "step": 15466
93921
+ },
93922
+ {
93923
+ "epoch": 123.29,
93924
+ "learning_rate": 9.766381260096931e-06,
93925
+ "loss": 0.4287,
93926
+ "step": 15467
93927
+ },
93928
+ {
93929
+ "epoch": 123.3,
93930
+ "learning_rate": 9.766365105008078e-06,
93931
+ "loss": 0.4472,
93932
+ "step": 15468
93933
+ },
93934
+ {
93935
+ "epoch": 123.3,
93936
+ "learning_rate": 9.766348949919225e-06,
93937
+ "loss": 0.4336,
93938
+ "step": 15469
93939
+ },
93940
+ {
93941
+ "epoch": 123.31,
93942
+ "learning_rate": 9.766332794830372e-06,
93943
+ "loss": 0.46,
93944
+ "step": 15470
93945
+ },
93946
+ {
93947
+ "epoch": 123.32,
93948
+ "learning_rate": 9.76631663974152e-06,
93949
+ "loss": 0.6141,
93950
+ "step": 15471
93951
+ },
93952
+ {
93953
+ "epoch": 123.33,
93954
+ "learning_rate": 9.766300484652667e-06,
93955
+ "loss": 0.5539,
93956
+ "step": 15472
93957
+ },
93958
+ {
93959
+ "epoch": 123.34,
93960
+ "learning_rate": 9.766284329563812e-06,
93961
+ "loss": 0.5473,
93962
+ "step": 15473
93963
+ },
93964
+ {
93965
+ "epoch": 123.34,
93966
+ "learning_rate": 9.76626817447496e-06,
93967
+ "loss": 0.7842,
93968
+ "step": 15474
93969
+ },
93970
+ {
93971
+ "epoch": 123.35,
93972
+ "learning_rate": 9.766252019386108e-06,
93973
+ "loss": 0.8673,
93974
+ "step": 15475
93975
+ },
93976
+ {
93977
+ "epoch": 123.36,
93978
+ "learning_rate": 9.766235864297255e-06,
93979
+ "loss": 0.8586,
93980
+ "step": 15476
93981
+ },
93982
+ {
93983
+ "epoch": 123.37,
93984
+ "learning_rate": 9.766219709208401e-06,
93985
+ "loss": 1.2058,
93986
+ "step": 15477
93987
+ },
93988
+ {
93989
+ "epoch": 123.38,
93990
+ "learning_rate": 9.766203554119548e-06,
93991
+ "loss": 1.4301,
93992
+ "step": 15478
93993
+ },
93994
+ {
93995
+ "epoch": 123.38,
93996
+ "learning_rate": 9.766187399030695e-06,
93997
+ "loss": 1.2956,
93998
+ "step": 15479
93999
+ },
94000
+ {
94001
+ "epoch": 123.39,
94002
+ "learning_rate": 9.766171243941842e-06,
94003
+ "loss": 2.0161,
94004
+ "step": 15480
94005
+ },
94006
+ {
94007
+ "epoch": 123.4,
94008
+ "learning_rate": 9.76615508885299e-06,
94009
+ "loss": 2.2544,
94010
+ "step": 15481
94011
+ },
94012
+ {
94013
+ "epoch": 123.41,
94014
+ "learning_rate": 9.766138933764137e-06,
94015
+ "loss": 0.692,
94016
+ "step": 15482
94017
+ },
94018
+ {
94019
+ "epoch": 123.42,
94020
+ "learning_rate": 9.766122778675282e-06,
94021
+ "loss": 0.6684,
94022
+ "step": 15483
94023
+ },
94024
+ {
94025
+ "epoch": 123.42,
94026
+ "learning_rate": 9.766106623586431e-06,
94027
+ "loss": 0.51,
94028
+ "step": 15484
94029
+ },
94030
+ {
94031
+ "epoch": 123.43,
94032
+ "learning_rate": 9.766090468497578e-06,
94033
+ "loss": 0.5415,
94034
+ "step": 15485
94035
+ },
94036
+ {
94037
+ "epoch": 123.44,
94038
+ "learning_rate": 9.766074313408725e-06,
94039
+ "loss": 1.1719,
94040
+ "step": 15486
94041
+ },
94042
+ {
94043
+ "epoch": 123.45,
94044
+ "learning_rate": 9.766058158319871e-06,
94045
+ "loss": 0.5499,
94046
+ "step": 15487
94047
+ },
94048
+ {
94049
+ "epoch": 123.46,
94050
+ "learning_rate": 9.766042003231018e-06,
94051
+ "loss": 0.4429,
94052
+ "step": 15488
94053
+ },
94054
+ {
94055
+ "epoch": 123.46,
94056
+ "learning_rate": 9.766025848142165e-06,
94057
+ "loss": 0.5409,
94058
+ "step": 15489
94059
+ },
94060
+ {
94061
+ "epoch": 123.47,
94062
+ "learning_rate": 9.766009693053312e-06,
94063
+ "loss": 0.5574,
94064
+ "step": 15490
94065
+ },
94066
+ {
94067
+ "epoch": 123.48,
94068
+ "learning_rate": 9.76599353796446e-06,
94069
+ "loss": 0.4656,
94070
+ "step": 15491
94071
+ },
94072
+ {
94073
+ "epoch": 123.49,
94074
+ "learning_rate": 9.765977382875607e-06,
94075
+ "loss": 0.5184,
94076
+ "step": 15492
94077
+ },
94078
+ {
94079
+ "epoch": 123.5,
94080
+ "learning_rate": 9.765961227786754e-06,
94081
+ "loss": 0.5391,
94082
+ "step": 15493
94083
+ },
94084
+ {
94085
+ "epoch": 123.5,
94086
+ "learning_rate": 9.765945072697901e-06,
94087
+ "loss": 0.606,
94088
+ "step": 15494
94089
+ },
94090
+ {
94091
+ "epoch": 123.51,
94092
+ "learning_rate": 9.765928917609048e-06,
94093
+ "loss": 0.6272,
94094
+ "step": 15495
94095
+ },
94096
+ {
94097
+ "epoch": 123.52,
94098
+ "learning_rate": 9.765912762520195e-06,
94099
+ "loss": 0.5646,
94100
+ "step": 15496
94101
+ },
94102
+ {
94103
+ "epoch": 123.53,
94104
+ "learning_rate": 9.765896607431341e-06,
94105
+ "loss": 0.766,
94106
+ "step": 15497
94107
+ },
94108
+ {
94109
+ "epoch": 123.54,
94110
+ "learning_rate": 9.765880452342488e-06,
94111
+ "loss": 0.5778,
94112
+ "step": 15498
94113
+ },
94114
+ {
94115
+ "epoch": 123.54,
94116
+ "learning_rate": 9.765864297253635e-06,
94117
+ "loss": 0.6132,
94118
+ "step": 15499
94119
+ },
94120
+ {
94121
+ "epoch": 123.55,
94122
+ "learning_rate": 9.765848142164782e-06,
94123
+ "loss": 0.744,
94124
+ "step": 15500
94125
+ },
94126
+ {
94127
+ "epoch": 123.56,
94128
+ "learning_rate": 9.76583198707593e-06,
94129
+ "loss": 1.4377,
94130
+ "step": 15501
94131
+ },
94132
+ {
94133
+ "epoch": 123.57,
94134
+ "learning_rate": 9.765815831987077e-06,
94135
+ "loss": 1.2746,
94136
+ "step": 15502
94137
+ },
94138
+ {
94139
+ "epoch": 123.58,
94140
+ "learning_rate": 9.765799676898224e-06,
94141
+ "loss": 1.475,
94142
+ "step": 15503
94143
+ },
94144
+ {
94145
+ "epoch": 123.58,
94146
+ "learning_rate": 9.765783521809371e-06,
94147
+ "loss": 1.5157,
94148
+ "step": 15504
94149
+ },
94150
+ {
94151
+ "epoch": 123.59,
94152
+ "learning_rate": 9.765767366720518e-06,
94153
+ "loss": 1.8992,
94154
+ "step": 15505
94155
+ },
94156
+ {
94157
+ "epoch": 123.6,
94158
+ "learning_rate": 9.765751211631665e-06,
94159
+ "loss": 3.2027,
94160
+ "step": 15506
94161
+ },
94162
+ {
94163
+ "epoch": 123.61,
94164
+ "learning_rate": 9.765735056542811e-06,
94165
+ "loss": 0.6823,
94166
+ "step": 15507
94167
+ },
94168
+ {
94169
+ "epoch": 123.62,
94170
+ "learning_rate": 9.765718901453958e-06,
94171
+ "loss": 0.5789,
94172
+ "step": 15508
94173
+ },
94174
+ {
94175
+ "epoch": 123.62,
94176
+ "learning_rate": 9.765702746365105e-06,
94177
+ "loss": 0.5084,
94178
+ "step": 15509
94179
+ },
94180
+ {
94181
+ "epoch": 123.63,
94182
+ "learning_rate": 9.765686591276254e-06,
94183
+ "loss": 0.7953,
94184
+ "step": 15510
94185
+ },
94186
+ {
94187
+ "epoch": 123.64,
94188
+ "learning_rate": 9.7656704361874e-06,
94189
+ "loss": 0.5224,
94190
+ "step": 15511
94191
+ },
94192
+ {
94193
+ "epoch": 123.65,
94194
+ "learning_rate": 9.765654281098547e-06,
94195
+ "loss": 0.8124,
94196
+ "step": 15512
94197
+ },
94198
+ {
94199
+ "epoch": 123.66,
94200
+ "learning_rate": 9.765638126009694e-06,
94201
+ "loss": 0.5162,
94202
+ "step": 15513
94203
+ },
94204
+ {
94205
+ "epoch": 123.66,
94206
+ "learning_rate": 9.765621970920841e-06,
94207
+ "loss": 0.3992,
94208
+ "step": 15514
94209
+ },
94210
+ {
94211
+ "epoch": 123.67,
94212
+ "learning_rate": 9.765605815831988e-06,
94213
+ "loss": 0.6249,
94214
+ "step": 15515
94215
+ },
94216
+ {
94217
+ "epoch": 123.68,
94218
+ "learning_rate": 9.765589660743135e-06,
94219
+ "loss": 0.4244,
94220
+ "step": 15516
94221
+ },
94222
+ {
94223
+ "epoch": 123.69,
94224
+ "learning_rate": 9.765573505654281e-06,
94225
+ "loss": 0.6223,
94226
+ "step": 15517
94227
+ },
94228
+ {
94229
+ "epoch": 123.7,
94230
+ "learning_rate": 9.765557350565428e-06,
94231
+ "loss": 0.5175,
94232
+ "step": 15518
94233
+ },
94234
+ {
94235
+ "epoch": 123.7,
94236
+ "learning_rate": 9.765541195476575e-06,
94237
+ "loss": 0.5203,
94238
+ "step": 15519
94239
+ },
94240
+ {
94241
+ "epoch": 123.71,
94242
+ "learning_rate": 9.765525040387724e-06,
94243
+ "loss": 0.7915,
94244
+ "step": 15520
94245
+ },
94246
+ {
94247
+ "epoch": 123.72,
94248
+ "learning_rate": 9.76550888529887e-06,
94249
+ "loss": 0.5749,
94250
+ "step": 15521
94251
+ },
94252
+ {
94253
+ "epoch": 123.73,
94254
+ "learning_rate": 9.765492730210017e-06,
94255
+ "loss": 0.5904,
94256
+ "step": 15522
94257
+ },
94258
+ {
94259
+ "epoch": 123.74,
94260
+ "learning_rate": 9.765476575121164e-06,
94261
+ "loss": 0.8081,
94262
+ "step": 15523
94263
+ },
94264
+ {
94265
+ "epoch": 123.74,
94266
+ "learning_rate": 9.76546042003231e-06,
94267
+ "loss": 0.9364,
94268
+ "step": 15524
94269
+ },
94270
+ {
94271
+ "epoch": 123.75,
94272
+ "learning_rate": 9.765444264943458e-06,
94273
+ "loss": 1.0647,
94274
+ "step": 15525
94275
+ },
94276
+ {
94277
+ "epoch": 123.76,
94278
+ "learning_rate": 9.765428109854606e-06,
94279
+ "loss": 0.8398,
94280
+ "step": 15526
94281
+ },
94282
+ {
94283
+ "epoch": 123.77,
94284
+ "learning_rate": 9.765411954765753e-06,
94285
+ "loss": 1.2186,
94286
+ "step": 15527
94287
+ },
94288
+ {
94289
+ "epoch": 123.78,
94290
+ "learning_rate": 9.765395799676898e-06,
94291
+ "loss": 1.0851,
94292
+ "step": 15528
94293
+ },
94294
+ {
94295
+ "epoch": 123.78,
94296
+ "learning_rate": 9.765379644588045e-06,
94297
+ "loss": 1.3124,
94298
+ "step": 15529
94299
+ },
94300
+ {
94301
+ "epoch": 123.79,
94302
+ "learning_rate": 9.765363489499194e-06,
94303
+ "loss": 1.966,
94304
+ "step": 15530
94305
+ },
94306
+ {
94307
+ "epoch": 123.8,
94308
+ "learning_rate": 9.76534733441034e-06,
94309
+ "loss": 2.3459,
94310
+ "step": 15531
94311
+ },
94312
+ {
94313
+ "epoch": 123.81,
94314
+ "learning_rate": 9.765331179321487e-06,
94315
+ "loss": 0.627,
94316
+ "step": 15532
94317
+ },
94318
+ {
94319
+ "epoch": 123.82,
94320
+ "learning_rate": 9.765315024232634e-06,
94321
+ "loss": 1.9344,
94322
+ "step": 15533
94323
+ },
94324
+ {
94325
+ "epoch": 123.82,
94326
+ "learning_rate": 9.76529886914378e-06,
94327
+ "loss": 0.4441,
94328
+ "step": 15534
94329
+ },
94330
+ {
94331
+ "epoch": 123.83,
94332
+ "learning_rate": 9.765282714054928e-06,
94333
+ "loss": 0.5598,
94334
+ "step": 15535
94335
+ },
94336
+ {
94337
+ "epoch": 123.84,
94338
+ "learning_rate": 9.765266558966076e-06,
94339
+ "loss": 0.5934,
94340
+ "step": 15536
94341
+ },
94342
+ {
94343
+ "epoch": 123.85,
94344
+ "learning_rate": 9.765250403877223e-06,
94345
+ "loss": 0.3741,
94346
+ "step": 15537
94347
+ },
94348
+ {
94349
+ "epoch": 123.86,
94350
+ "learning_rate": 9.765234248788368e-06,
94351
+ "loss": 0.4276,
94352
+ "step": 15538
94353
+ },
94354
+ {
94355
+ "epoch": 123.86,
94356
+ "learning_rate": 9.765218093699517e-06,
94357
+ "loss": 3.0023,
94358
+ "step": 15539
94359
+ },
94360
+ {
94361
+ "epoch": 123.87,
94362
+ "learning_rate": 9.765201938610663e-06,
94363
+ "loss": 0.4582,
94364
+ "step": 15540
94365
+ },
94366
+ {
94367
+ "epoch": 123.88,
94368
+ "learning_rate": 9.76518578352181e-06,
94369
+ "loss": 0.468,
94370
+ "step": 15541
94371
+ },
94372
+ {
94373
+ "epoch": 123.89,
94374
+ "learning_rate": 9.765169628432957e-06,
94375
+ "loss": 0.4791,
94376
+ "step": 15542
94377
+ },
94378
+ {
94379
+ "epoch": 123.9,
94380
+ "learning_rate": 9.765153473344104e-06,
94381
+ "loss": 0.5,
94382
+ "step": 15543
94383
+ },
94384
+ {
94385
+ "epoch": 123.9,
94386
+ "learning_rate": 9.76513731825525e-06,
94387
+ "loss": 0.5137,
94388
+ "step": 15544
94389
+ },
94390
+ {
94391
+ "epoch": 123.91,
94392
+ "learning_rate": 9.765121163166398e-06,
94393
+ "loss": 0.4238,
94394
+ "step": 15545
94395
+ },
94396
+ {
94397
+ "epoch": 123.92,
94398
+ "learning_rate": 9.765105008077546e-06,
94399
+ "loss": 0.4912,
94400
+ "step": 15546
94401
+ },
94402
+ {
94403
+ "epoch": 123.93,
94404
+ "learning_rate": 9.765088852988693e-06,
94405
+ "loss": 0.8465,
94406
+ "step": 15547
94407
+ },
94408
+ {
94409
+ "epoch": 123.94,
94410
+ "learning_rate": 9.765072697899838e-06,
94411
+ "loss": 0.6005,
94412
+ "step": 15548
94413
+ },
94414
+ {
94415
+ "epoch": 123.94,
94416
+ "learning_rate": 9.765056542810987e-06,
94417
+ "loss": 0.9715,
94418
+ "step": 15549
94419
+ },
94420
+ {
94421
+ "epoch": 123.95,
94422
+ "learning_rate": 9.765040387722133e-06,
94423
+ "loss": 0.7747,
94424
+ "step": 15550
94425
+ },
94426
+ {
94427
+ "epoch": 123.96,
94428
+ "learning_rate": 9.76502423263328e-06,
94429
+ "loss": 0.995,
94430
+ "step": 15551
94431
+ },
94432
+ {
94433
+ "epoch": 123.97,
94434
+ "learning_rate": 9.765008077544427e-06,
94435
+ "loss": 1.073,
94436
+ "step": 15552
94437
+ },
94438
+ {
94439
+ "epoch": 123.98,
94440
+ "learning_rate": 9.764991922455574e-06,
94441
+ "loss": 1.1276,
94442
+ "step": 15553
94443
+ },
94444
+ {
94445
+ "epoch": 123.98,
94446
+ "learning_rate": 9.76497576736672e-06,
94447
+ "loss": 1.3905,
94448
+ "step": 15554
94449
+ },
94450
+ {
94451
+ "epoch": 123.99,
94452
+ "learning_rate": 9.764959612277868e-06,
94453
+ "loss": 2.1978,
94454
+ "step": 15555
94455
+ },
94456
+ {
94457
+ "epoch": 124.0,
94458
+ "learning_rate": 9.764943457189016e-06,
94459
+ "loss": 3.3244,
94460
+ "step": 15556
94461
+ },
94462
+ {
94463
+ "epoch": 124.0,
94464
+ "eval_loss": 0.6700283288955688,
94465
+ "eval_runtime": 38.4234,
94466
+ "eval_samples_per_second": 21.784,
94467
+ "eval_steps_per_second": 0.703,
94468
+ "eval_wer": 0.27634520792964673,
94469
+ "step": 15556
94470
+ },
94471
+ {
94472
+ "epoch": 125.01,
94473
+ "learning_rate": 9.764927302100163e-06,
94474
+ "loss": 0.8589,
94475
+ "step": 15557
94476
+ },
94477
+ {
94478
+ "epoch": 125.02,
94479
+ "learning_rate": 9.76491114701131e-06,
94480
+ "loss": 0.4849,
94481
+ "step": 15558
94482
+ },
94483
+ {
94484
+ "epoch": 125.02,
94485
+ "learning_rate": 9.764894991922457e-06,
94486
+ "loss": 0.5023,
94487
+ "step": 15559
94488
+ },
94489
+ {
94490
+ "epoch": 125.03,
94491
+ "learning_rate": 9.764878836833603e-06,
94492
+ "loss": 0.5933,
94493
+ "step": 15560
94494
+ },
94495
+ {
94496
+ "epoch": 125.04,
94497
+ "learning_rate": 9.76486268174475e-06,
94498
+ "loss": 0.484,
94499
+ "step": 15561
94500
+ },
94501
+ {
94502
+ "epoch": 125.05,
94503
+ "learning_rate": 9.764846526655897e-06,
94504
+ "loss": 0.5367,
94505
+ "step": 15562
94506
+ },
94507
+ {
94508
+ "epoch": 125.06,
94509
+ "learning_rate": 9.764830371567044e-06,
94510
+ "loss": 1.1475,
94511
+ "step": 15563
94512
+ },
94513
+ {
94514
+ "epoch": 125.06,
94515
+ "learning_rate": 9.76481421647819e-06,
94516
+ "loss": 0.5011,
94517
+ "step": 15564
94518
+ },
94519
+ {
94520
+ "epoch": 125.07,
94521
+ "learning_rate": 9.76479806138934e-06,
94522
+ "loss": 0.4924,
94523
+ "step": 15565
94524
+ },
94525
+ {
94526
+ "epoch": 125.08,
94527
+ "learning_rate": 9.764781906300486e-06,
94528
+ "loss": 0.6192,
94529
+ "step": 15566
94530
+ },
94531
+ {
94532
+ "epoch": 125.09,
94533
+ "learning_rate": 9.764765751211633e-06,
94534
+ "loss": 0.4957,
94535
+ "step": 15567
94536
+ },
94537
+ {
94538
+ "epoch": 125.1,
94539
+ "learning_rate": 9.76474959612278e-06,
94540
+ "loss": 0.6133,
94541
+ "step": 15568
94542
+ },
94543
+ {
94544
+ "epoch": 125.1,
94545
+ "learning_rate": 9.764733441033927e-06,
94546
+ "loss": 0.7008,
94547
+ "step": 15569
94548
+ },
94549
+ {
94550
+ "epoch": 125.11,
94551
+ "learning_rate": 9.764717285945073e-06,
94552
+ "loss": 0.4287,
94553
+ "step": 15570
94554
+ },
94555
+ {
94556
+ "epoch": 125.12,
94557
+ "learning_rate": 9.76470113085622e-06,
94558
+ "loss": 0.685,
94559
+ "step": 15571
94560
+ },
94561
+ {
94562
+ "epoch": 125.13,
94563
+ "learning_rate": 9.764684975767367e-06,
94564
+ "loss": 0.8954,
94565
+ "step": 15572
94566
+ },
94567
+ {
94568
+ "epoch": 125.14,
94569
+ "learning_rate": 9.764668820678514e-06,
94570
+ "loss": 0.7584,
94571
+ "step": 15573
94572
+ },
94573
+ {
94574
+ "epoch": 125.14,
94575
+ "learning_rate": 9.76465266558966e-06,
94576
+ "loss": 1.1585,
94577
+ "step": 15574
94578
+ },
94579
+ {
94580
+ "epoch": 125.15,
94581
+ "learning_rate": 9.76463651050081e-06,
94582
+ "loss": 1.5004,
94583
+ "step": 15575
94584
+ },
94585
+ {
94586
+ "epoch": 125.16,
94587
+ "learning_rate": 9.764620355411956e-06,
94588
+ "loss": 1.013,
94589
+ "step": 15576
94590
+ },
94591
+ {
94592
+ "epoch": 125.17,
94593
+ "learning_rate": 9.764604200323103e-06,
94594
+ "loss": 1.0403,
94595
+ "step": 15577
94596
+ },
94597
+ {
94598
+ "epoch": 125.18,
94599
+ "learning_rate": 9.76458804523425e-06,
94600
+ "loss": 1.295,
94601
+ "step": 15578
94602
+ },
94603
+ {
94604
+ "epoch": 125.18,
94605
+ "learning_rate": 9.764571890145397e-06,
94606
+ "loss": 1.717,
94607
+ "step": 15579
94608
+ },
94609
+ {
94610
+ "epoch": 125.19,
94611
+ "learning_rate": 9.764555735056543e-06,
94612
+ "loss": 1.8107,
94613
+ "step": 15580
94614
+ },
94615
+ {
94616
+ "epoch": 125.2,
94617
+ "learning_rate": 9.76453957996769e-06,
94618
+ "loss": 2.327,
94619
+ "step": 15581
94620
+ },
94621
+ {
94622
+ "epoch": 125.21,
94623
+ "learning_rate": 9.764523424878837e-06,
94624
+ "loss": 0.5922,
94625
+ "step": 15582
94626
+ },
94627
+ {
94628
+ "epoch": 125.22,
94629
+ "learning_rate": 9.764507269789984e-06,
94630
+ "loss": 0.5566,
94631
+ "step": 15583
94632
+ },
94633
+ {
94634
+ "epoch": 125.22,
94635
+ "learning_rate": 9.76449111470113e-06,
94636
+ "loss": 0.728,
94637
+ "step": 15584
94638
+ },
94639
+ {
94640
+ "epoch": 125.23,
94641
+ "learning_rate": 9.76447495961228e-06,
94642
+ "loss": 0.5909,
94643
+ "step": 15585
94644
+ },
94645
+ {
94646
+ "epoch": 125.24,
94647
+ "learning_rate": 9.764458804523426e-06,
94648
+ "loss": 0.468,
94649
+ "step": 15586
94650
+ },
94651
+ {
94652
+ "epoch": 125.25,
94653
+ "learning_rate": 9.764442649434573e-06,
94654
+ "loss": 0.4003,
94655
+ "step": 15587
94656
+ },
94657
+ {
94658
+ "epoch": 125.26,
94659
+ "learning_rate": 9.76442649434572e-06,
94660
+ "loss": 0.3779,
94661
+ "step": 15588
94662
+ },
94663
+ {
94664
+ "epoch": 125.27,
94665
+ "learning_rate": 9.764410339256867e-06,
94666
+ "loss": 0.487,
94667
+ "step": 15589
94668
+ },
94669
+ {
94670
+ "epoch": 125.27,
94671
+ "learning_rate": 9.764394184168013e-06,
94672
+ "loss": 0.486,
94673
+ "step": 15590
94674
+ },
94675
+ {
94676
+ "epoch": 125.28,
94677
+ "learning_rate": 9.764378029079162e-06,
94678
+ "loss": 0.5336,
94679
+ "step": 15591
94680
+ },
94681
+ {
94682
+ "epoch": 125.29,
94683
+ "learning_rate": 9.764361873990309e-06,
94684
+ "loss": 0.5435,
94685
+ "step": 15592
94686
+ },
94687
+ {
94688
+ "epoch": 125.3,
94689
+ "learning_rate": 9.764345718901454e-06,
94690
+ "loss": 0.6432,
94691
+ "step": 15593
94692
+ },
94693
+ {
94694
+ "epoch": 125.31,
94695
+ "learning_rate": 9.7643295638126e-06,
94696
+ "loss": 0.5737,
94697
+ "step": 15594
94698
+ },
94699
+ {
94700
+ "epoch": 125.31,
94701
+ "learning_rate": 9.76431340872375e-06,
94702
+ "loss": 0.5901,
94703
+ "step": 15595
94704
+ },
94705
+ {
94706
+ "epoch": 125.32,
94707
+ "learning_rate": 9.764297253634896e-06,
94708
+ "loss": 0.648,
94709
+ "step": 15596
94710
+ },
94711
+ {
94712
+ "epoch": 125.33,
94713
+ "learning_rate": 9.764281098546043e-06,
94714
+ "loss": 0.5778,
94715
+ "step": 15597
94716
+ },
94717
+ {
94718
+ "epoch": 125.34,
94719
+ "learning_rate": 9.76426494345719e-06,
94720
+ "loss": 0.7171,
94721
+ "step": 15598
94722
+ },
94723
+ {
94724
+ "epoch": 125.35,
94725
+ "learning_rate": 9.764248788368337e-06,
94726
+ "loss": 0.9377,
94727
+ "step": 15599
94728
+ },
94729
+ {
94730
+ "epoch": 125.35,
94731
+ "learning_rate": 9.764232633279483e-06,
94732
+ "loss": 0.7668,
94733
+ "step": 15600
94734
+ },
94735
+ {
94736
+ "epoch": 125.36,
94737
+ "learning_rate": 9.764216478190632e-06,
94738
+ "loss": 0.9051,
94739
+ "step": 15601
94740
+ },
94741
+ {
94742
+ "epoch": 125.37,
94743
+ "learning_rate": 9.764200323101779e-06,
94744
+ "loss": 1.0389,
94745
+ "step": 15602
94746
+ },
94747
+ {
94748
+ "epoch": 125.38,
94749
+ "learning_rate": 9.764184168012924e-06,
94750
+ "loss": 1.1533,
94751
+ "step": 15603
94752
+ },
94753
+ {
94754
+ "epoch": 125.39,
94755
+ "learning_rate": 9.764168012924072e-06,
94756
+ "loss": 1.329,
94757
+ "step": 15604
94758
+ },
94759
+ {
94760
+ "epoch": 125.39,
94761
+ "learning_rate": 9.76415185783522e-06,
94762
+ "loss": 1.6055,
94763
+ "step": 15605
94764
+ },
94765
+ {
94766
+ "epoch": 125.4,
94767
+ "learning_rate": 9.764135702746366e-06,
94768
+ "loss": 2.3053,
94769
+ "step": 15606
94770
+ },
94771
+ {
94772
+ "epoch": 125.41,
94773
+ "learning_rate": 9.764119547657513e-06,
94774
+ "loss": 0.568,
94775
+ "step": 15607
94776
+ },
94777
+ {
94778
+ "epoch": 125.42,
94779
+ "learning_rate": 9.76410339256866e-06,
94780
+ "loss": 0.577,
94781
+ "step": 15608
94782
+ },
94783
+ {
94784
+ "epoch": 125.43,
94785
+ "learning_rate": 9.764087237479806e-06,
94786
+ "loss": 0.7538,
94787
+ "step": 15609
94788
+ },
94789
+ {
94790
+ "epoch": 125.43,
94791
+ "learning_rate": 9.764071082390953e-06,
94792
+ "loss": 0.4809,
94793
+ "step": 15610
94794
+ },
94795
+ {
94796
+ "epoch": 125.44,
94797
+ "learning_rate": 9.764054927302102e-06,
94798
+ "loss": 0.4842,
94799
+ "step": 15611
94800
+ },
94801
+ {
94802
+ "epoch": 125.45,
94803
+ "learning_rate": 9.764038772213249e-06,
94804
+ "loss": 0.4789,
94805
+ "step": 15612
94806
+ },
94807
+ {
94808
+ "epoch": 125.46,
94809
+ "learning_rate": 9.764022617124394e-06,
94810
+ "loss": 0.5795,
94811
+ "step": 15613
94812
+ },
94813
+ {
94814
+ "epoch": 125.47,
94815
+ "learning_rate": 9.764006462035542e-06,
94816
+ "loss": 0.6214,
94817
+ "step": 15614
94818
+ },
94819
+ {
94820
+ "epoch": 125.47,
94821
+ "learning_rate": 9.763990306946689e-06,
94822
+ "loss": 0.9528,
94823
+ "step": 15615
94824
+ },
94825
+ {
94826
+ "epoch": 125.48,
94827
+ "learning_rate": 9.763974151857836e-06,
94828
+ "loss": 0.4921,
94829
+ "step": 15616
94830
+ },
94831
+ {
94832
+ "epoch": 125.49,
94833
+ "learning_rate": 9.763957996768983e-06,
94834
+ "loss": 0.4861,
94835
+ "step": 15617
94836
+ },
94837
+ {
94838
+ "epoch": 125.5,
94839
+ "learning_rate": 9.76394184168013e-06,
94840
+ "loss": 0.5151,
94841
+ "step": 15618
94842
+ },
94843
+ {
94844
+ "epoch": 125.51,
94845
+ "learning_rate": 9.763925686591276e-06,
94846
+ "loss": 0.4528,
94847
+ "step": 15619
94848
+ },
94849
+ {
94850
+ "epoch": 125.51,
94851
+ "learning_rate": 9.763909531502425e-06,
94852
+ "loss": 0.649,
94853
+ "step": 15620
94854
+ },
94855
+ {
94856
+ "epoch": 125.52,
94857
+ "learning_rate": 9.763893376413572e-06,
94858
+ "loss": 0.558,
94859
+ "step": 15621
94860
+ },
94861
+ {
94862
+ "epoch": 125.53,
94863
+ "learning_rate": 9.763877221324719e-06,
94864
+ "loss": 0.5238,
94865
+ "step": 15622
94866
+ },
94867
+ {
94868
+ "epoch": 125.54,
94869
+ "learning_rate": 9.763861066235864e-06,
94870
+ "loss": 0.6497,
94871
+ "step": 15623
94872
+ },
94873
+ {
94874
+ "epoch": 125.55,
94875
+ "learning_rate": 9.763844911147012e-06,
94876
+ "loss": 0.6721,
94877
+ "step": 15624
94878
+ },
94879
+ {
94880
+ "epoch": 125.55,
94881
+ "learning_rate": 9.763828756058159e-06,
94882
+ "loss": 0.7368,
94883
+ "step": 15625
94884
+ },
94885
+ {
94886
+ "epoch": 125.56,
94887
+ "learning_rate": 9.763812600969306e-06,
94888
+ "loss": 1.1011,
94889
+ "step": 15626
94890
+ },
94891
+ {
94892
+ "epoch": 125.57,
94893
+ "learning_rate": 9.763796445880453e-06,
94894
+ "loss": 1.0173,
94895
+ "step": 15627
94896
+ },
94897
+ {
94898
+ "epoch": 125.58,
94899
+ "learning_rate": 9.7637802907916e-06,
94900
+ "loss": 1.2385,
94901
+ "step": 15628
94902
+ },
94903
+ {
94904
+ "epoch": 125.59,
94905
+ "learning_rate": 9.763764135702746e-06,
94906
+ "loss": 1.4176,
94907
+ "step": 15629
94908
+ },
94909
+ {
94910
+ "epoch": 125.59,
94911
+ "learning_rate": 9.763747980613895e-06,
94912
+ "loss": 1.7172,
94913
+ "step": 15630
94914
+ },
94915
+ {
94916
+ "epoch": 125.6,
94917
+ "learning_rate": 9.763731825525042e-06,
94918
+ "loss": 2.1914,
94919
+ "step": 15631
94920
+ },
94921
+ {
94922
+ "epoch": 125.61,
94923
+ "learning_rate": 9.763715670436189e-06,
94924
+ "loss": 0.7761,
94925
+ "step": 15632
94926
+ },
94927
+ {
94928
+ "epoch": 125.62,
94929
+ "learning_rate": 9.763699515347335e-06,
94930
+ "loss": 0.8653,
94931
+ "step": 15633
94932
+ },
94933
+ {
94934
+ "epoch": 125.63,
94935
+ "learning_rate": 9.763683360258482e-06,
94936
+ "loss": 0.6948,
94937
+ "step": 15634
94938
+ },
94939
+ {
94940
+ "epoch": 125.63,
94941
+ "learning_rate": 9.763667205169629e-06,
94942
+ "loss": 0.4559,
94943
+ "step": 15635
94944
+ },
94945
+ {
94946
+ "epoch": 125.64,
94947
+ "learning_rate": 9.763651050080776e-06,
94948
+ "loss": 0.517,
94949
+ "step": 15636
94950
+ },
94951
+ {
94952
+ "epoch": 125.65,
94953
+ "learning_rate": 9.763634894991923e-06,
94954
+ "loss": 0.4403,
94955
+ "step": 15637
94956
+ },
94957
+ {
94958
+ "epoch": 125.66,
94959
+ "learning_rate": 9.76361873990307e-06,
94960
+ "loss": 0.4218,
94961
+ "step": 15638
94962
+ },
94963
+ {
94964
+ "epoch": 125.67,
94965
+ "learning_rate": 9.763602584814216e-06,
94966
+ "loss": 0.4505,
94967
+ "step": 15639
94968
+ },
94969
+ {
94970
+ "epoch": 125.67,
94971
+ "learning_rate": 9.763586429725365e-06,
94972
+ "loss": 0.683,
94973
+ "step": 15640
94974
+ },
94975
+ {
94976
+ "epoch": 125.68,
94977
+ "learning_rate": 9.763570274636512e-06,
94978
+ "loss": 0.5036,
94979
+ "step": 15641
94980
+ },
94981
+ {
94982
+ "epoch": 125.69,
94983
+ "learning_rate": 9.763554119547659e-06,
94984
+ "loss": 0.4833,
94985
+ "step": 15642
94986
+ },
94987
+ {
94988
+ "epoch": 125.7,
94989
+ "learning_rate": 9.763537964458805e-06,
94990
+ "loss": 0.474,
94991
+ "step": 15643
94992
+ },
94993
+ {
94994
+ "epoch": 125.71,
94995
+ "learning_rate": 9.763521809369952e-06,
94996
+ "loss": 0.5592,
94997
+ "step": 15644
94998
+ },
94999
+ {
95000
+ "epoch": 125.71,
95001
+ "learning_rate": 9.763505654281099e-06,
95002
+ "loss": 0.4352,
95003
+ "step": 15645
95004
+ },
95005
+ {
95006
+ "epoch": 125.72,
95007
+ "learning_rate": 9.763489499192248e-06,
95008
+ "loss": 0.6165,
95009
+ "step": 15646
95010
+ },
95011
+ {
95012
+ "epoch": 125.73,
95013
+ "learning_rate": 9.763473344103393e-06,
95014
+ "loss": 0.5416,
95015
+ "step": 15647
95016
+ },
95017
+ {
95018
+ "epoch": 125.74,
95019
+ "learning_rate": 9.76345718901454e-06,
95020
+ "loss": 0.6219,
95021
+ "step": 15648
95022
+ },
95023
+ {
95024
+ "epoch": 125.75,
95025
+ "learning_rate": 9.763441033925686e-06,
95026
+ "loss": 0.7089,
95027
+ "step": 15649
95028
+ },
95029
+ {
95030
+ "epoch": 125.76,
95031
+ "learning_rate": 9.763424878836835e-06,
95032
+ "loss": 0.9465,
95033
+ "step": 15650
95034
+ },
95035
+ {
95036
+ "epoch": 125.76,
95037
+ "learning_rate": 9.763408723747982e-06,
95038
+ "loss": 1.0269,
95039
+ "step": 15651
95040
+ },
95041
+ {
95042
+ "epoch": 125.77,
95043
+ "learning_rate": 9.763392568659129e-06,
95044
+ "loss": 1.0963,
95045
+ "step": 15652
95046
+ },
95047
+ {
95048
+ "epoch": 125.78,
95049
+ "learning_rate": 9.763376413570275e-06,
95050
+ "loss": 1.0794,
95051
+ "step": 15653
95052
+ },
95053
+ {
95054
+ "epoch": 125.79,
95055
+ "learning_rate": 9.763360258481422e-06,
95056
+ "loss": 1.3015,
95057
+ "step": 15654
95058
+ },
95059
+ {
95060
+ "epoch": 125.8,
95061
+ "learning_rate": 9.763344103392569e-06,
95062
+ "loss": 1.9667,
95063
+ "step": 15655
95064
+ },
95065
+ {
95066
+ "epoch": 125.8,
95067
+ "learning_rate": 9.763327948303718e-06,
95068
+ "loss": 1.9323,
95069
+ "step": 15656
95070
+ },
95071
+ {
95072
+ "epoch": 125.81,
95073
+ "learning_rate": 9.763311793214864e-06,
95074
+ "loss": 0.8994,
95075
+ "step": 15657
95076
+ },
95077
+ {
95078
+ "epoch": 125.82,
95079
+ "learning_rate": 9.76329563812601e-06,
95080
+ "loss": 0.5849,
95081
+ "step": 15658
95082
+ },
95083
+ {
95084
+ "epoch": 125.83,
95085
+ "learning_rate": 9.763279483037158e-06,
95086
+ "loss": 0.4049,
95087
+ "step": 15659
95088
+ },
95089
+ {
95090
+ "epoch": 125.84,
95091
+ "learning_rate": 9.763263327948305e-06,
95092
+ "loss": 0.5683,
95093
+ "step": 15660
95094
+ },
95095
+ {
95096
+ "epoch": 125.84,
95097
+ "learning_rate": 9.763247172859452e-06,
95098
+ "loss": 0.4702,
95099
+ "step": 15661
95100
+ },
95101
+ {
95102
+ "epoch": 125.85,
95103
+ "learning_rate": 9.763231017770599e-06,
95104
+ "loss": 0.7488,
95105
+ "step": 15662
95106
+ },
95107
+ {
95108
+ "epoch": 125.86,
95109
+ "learning_rate": 9.763214862681745e-06,
95110
+ "loss": 0.4374,
95111
+ "step": 15663
95112
+ },
95113
+ {
95114
+ "epoch": 125.87,
95115
+ "learning_rate": 9.763198707592892e-06,
95116
+ "loss": 0.6319,
95117
+ "step": 15664
95118
+ },
95119
+ {
95120
+ "epoch": 125.88,
95121
+ "learning_rate": 9.763182552504039e-06,
95122
+ "loss": 0.5642,
95123
+ "step": 15665
95124
+ },
95125
+ {
95126
+ "epoch": 125.88,
95127
+ "learning_rate": 9.763166397415188e-06,
95128
+ "loss": 0.5154,
95129
+ "step": 15666
95130
+ },
95131
+ {
95132
+ "epoch": 125.89,
95133
+ "learning_rate": 9.763150242326334e-06,
95134
+ "loss": 0.6278,
95135
+ "step": 15667
95136
+ },
95137
+ {
95138
+ "epoch": 125.9,
95139
+ "learning_rate": 9.76313408723748e-06,
95140
+ "loss": 0.5362,
95141
+ "step": 15668
95142
+ },
95143
+ {
95144
+ "epoch": 125.91,
95145
+ "learning_rate": 9.763117932148628e-06,
95146
+ "loss": 0.8385,
95147
+ "step": 15669
95148
+ },
95149
+ {
95150
+ "epoch": 125.92,
95151
+ "learning_rate": 9.763101777059775e-06,
95152
+ "loss": 0.5664,
95153
+ "step": 15670
95154
+ },
95155
+ {
95156
+ "epoch": 125.92,
95157
+ "learning_rate": 9.763085621970922e-06,
95158
+ "loss": 0.5831,
95159
+ "step": 15671
95160
+ },
95161
+ {
95162
+ "epoch": 125.93,
95163
+ "learning_rate": 9.763069466882069e-06,
95164
+ "loss": 0.6457,
95165
+ "step": 15672
95166
+ },
95167
+ {
95168
+ "epoch": 125.94,
95169
+ "learning_rate": 9.763053311793215e-06,
95170
+ "loss": 0.7199,
95171
+ "step": 15673
95172
+ },
95173
+ {
95174
+ "epoch": 125.95,
95175
+ "learning_rate": 9.763037156704362e-06,
95176
+ "loss": 0.5475,
95177
+ "step": 15674
95178
+ },
95179
+ {
95180
+ "epoch": 125.96,
95181
+ "learning_rate": 9.763021001615509e-06,
95182
+ "loss": 1.0718,
95183
+ "step": 15675
95184
+ },
95185
+ {
95186
+ "epoch": 125.96,
95187
+ "learning_rate": 9.763004846526658e-06,
95188
+ "loss": 1.2341,
95189
+ "step": 15676
95190
+ },
95191
+ {
95192
+ "epoch": 125.97,
95193
+ "learning_rate": 9.762988691437804e-06,
95194
+ "loss": 1.0493,
95195
+ "step": 15677
95196
+ },
95197
+ {
95198
+ "epoch": 125.98,
95199
+ "learning_rate": 9.76297253634895e-06,
95200
+ "loss": 1.2414,
95201
+ "step": 15678
95202
+ },
95203
+ {
95204
+ "epoch": 125.99,
95205
+ "learning_rate": 9.762956381260098e-06,
95206
+ "loss": 1.4355,
95207
+ "step": 15679
95208
+ },
95209
+ {
95210
+ "epoch": 126.0,
95211
+ "learning_rate": 9.762940226171245e-06,
95212
+ "loss": 2.3495,
95213
+ "step": 15680
95214
+ },
95215
+ {
95216
+ "epoch": 126.0,
95217
+ "eval_loss": 0.6309465765953064,
95218
+ "eval_runtime": 40.4147,
95219
+ "eval_samples_per_second": 20.71,
95220
+ "eval_steps_per_second": 0.668,
95221
+ "eval_wer": 0.2643427741466957,
95222
+ "step": 15680
95223
+ },
95224
+ {
95225
+ "epoch": 126.01,
95226
+ "learning_rate": 9.762924071082392e-06,
95227
+ "loss": 1.0584,
95228
+ "step": 15681
95229
+ },
95230
+ {
95231
+ "epoch": 126.02,
95232
+ "learning_rate": 9.762907915993538e-06,
95233
+ "loss": 0.6019,
95234
+ "step": 15682
95235
+ },
95236
+ {
95237
+ "epoch": 126.02,
95238
+ "learning_rate": 9.762891760904685e-06,
95239
+ "loss": 0.5021,
95240
+ "step": 15683
95241
+ },
95242
+ {
95243
+ "epoch": 126.03,
95244
+ "learning_rate": 9.762875605815832e-06,
95245
+ "loss": 0.4821,
95246
+ "step": 15684
95247
+ },
95248
+ {
95249
+ "epoch": 126.04,
95250
+ "learning_rate": 9.76285945072698e-06,
95251
+ "loss": 0.4379,
95252
+ "step": 15685
95253
+ },
95254
+ {
95255
+ "epoch": 126.05,
95256
+ "learning_rate": 9.762843295638127e-06,
95257
+ "loss": 0.8002,
95258
+ "step": 15686
95259
+ },
95260
+ {
95261
+ "epoch": 126.06,
95262
+ "learning_rate": 9.762827140549274e-06,
95263
+ "loss": 0.5978,
95264
+ "step": 15687
95265
+ },
95266
+ {
95267
+ "epoch": 126.06,
95268
+ "learning_rate": 9.76281098546042e-06,
95269
+ "loss": 0.463,
95270
+ "step": 15688
95271
+ },
95272
+ {
95273
+ "epoch": 126.07,
95274
+ "learning_rate": 9.762794830371568e-06,
95275
+ "loss": 0.5929,
95276
+ "step": 15689
95277
+ },
95278
+ {
95279
+ "epoch": 126.08,
95280
+ "learning_rate": 9.762778675282715e-06,
95281
+ "loss": 0.5566,
95282
+ "step": 15690
95283
+ },
95284
+ {
95285
+ "epoch": 126.09,
95286
+ "learning_rate": 9.762762520193862e-06,
95287
+ "loss": 0.5328,
95288
+ "step": 15691
95289
+ },
95290
+ {
95291
+ "epoch": 126.1,
95292
+ "learning_rate": 9.762746365105008e-06,
95293
+ "loss": 0.7164,
95294
+ "step": 15692
95295
+ },
95296
+ {
95297
+ "epoch": 126.1,
95298
+ "learning_rate": 9.762730210016155e-06,
95299
+ "loss": 0.8021,
95300
+ "step": 15693
95301
+ },
95302
+ {
95303
+ "epoch": 126.11,
95304
+ "learning_rate": 9.762714054927302e-06,
95305
+ "loss": 0.7822,
95306
+ "step": 15694
95307
+ },
95308
+ {
95309
+ "epoch": 126.12,
95310
+ "learning_rate": 9.76269789983845e-06,
95311
+ "loss": 0.5048,
95312
+ "step": 15695
95313
+ },
95314
+ {
95315
+ "epoch": 126.13,
95316
+ "learning_rate": 9.762681744749597e-06,
95317
+ "loss": 0.5867,
95318
+ "step": 15696
95319
+ },
95320
+ {
95321
+ "epoch": 126.14,
95322
+ "learning_rate": 9.762665589660744e-06,
95323
+ "loss": 0.5806,
95324
+ "step": 15697
95325
+ },
95326
+ {
95327
+ "epoch": 126.14,
95328
+ "learning_rate": 9.762649434571891e-06,
95329
+ "loss": 0.7003,
95330
+ "step": 15698
95331
+ },
95332
+ {
95333
+ "epoch": 126.15,
95334
+ "learning_rate": 9.762633279483038e-06,
95335
+ "loss": 0.673,
95336
+ "step": 15699
95337
+ },
95338
+ {
95339
+ "epoch": 126.16,
95340
+ "learning_rate": 9.762617124394185e-06,
95341
+ "loss": 1.0922,
95342
+ "step": 15700
95343
+ },
95344
+ {
95345
+ "epoch": 126.17,
95346
+ "learning_rate": 9.762600969305333e-06,
95347
+ "loss": 0.9486,
95348
+ "step": 15701
95349
+ },
95350
+ {
95351
+ "epoch": 126.18,
95352
+ "learning_rate": 9.762584814216478e-06,
95353
+ "loss": 1.3478,
95354
+ "step": 15702
95355
+ },
95356
+ {
95357
+ "epoch": 126.18,
95358
+ "learning_rate": 9.762568659127625e-06,
95359
+ "loss": 1.4727,
95360
+ "step": 15703
95361
+ },
95362
+ {
95363
+ "epoch": 126.19,
95364
+ "learning_rate": 9.762552504038772e-06,
95365
+ "loss": 2.1602,
95366
+ "step": 15704
95367
+ },
95368
+ {
95369
+ "epoch": 126.2,
95370
+ "learning_rate": 9.76253634894992e-06,
95371
+ "loss": 2.4559,
95372
+ "step": 15705
95373
+ },
95374
+ {
95375
+ "epoch": 126.21,
95376
+ "learning_rate": 9.762520193861067e-06,
95377
+ "loss": 1.5962,
95378
+ "step": 15706
95379
+ },
95380
+ {
95381
+ "epoch": 126.22,
95382
+ "learning_rate": 9.762504038772214e-06,
95383
+ "loss": 0.8303,
95384
+ "step": 15707
95385
+ },
95386
+ {
95387
+ "epoch": 126.22,
95388
+ "learning_rate": 9.762487883683361e-06,
95389
+ "loss": 0.5114,
95390
+ "step": 15708
95391
+ },
95392
+ {
95393
+ "epoch": 126.23,
95394
+ "learning_rate": 9.762471728594508e-06,
95395
+ "loss": 0.8258,
95396
+ "step": 15709
95397
+ },
95398
+ {
95399
+ "epoch": 126.24,
95400
+ "learning_rate": 9.762455573505655e-06,
95401
+ "loss": 0.7666,
95402
+ "step": 15710
95403
+ },
95404
+ {
95405
+ "epoch": 126.25,
95406
+ "learning_rate": 9.762439418416803e-06,
95407
+ "loss": 0.5342,
95408
+ "step": 15711
95409
+ },
95410
+ {
95411
+ "epoch": 126.26,
95412
+ "learning_rate": 9.762423263327948e-06,
95413
+ "loss": 0.463,
95414
+ "step": 15712
95415
+ },
95416
+ {
95417
+ "epoch": 126.27,
95418
+ "learning_rate": 9.762407108239095e-06,
95419
+ "loss": 0.4412,
95420
+ "step": 15713
95421
+ },
95422
+ {
95423
+ "epoch": 126.27,
95424
+ "learning_rate": 9.762390953150244e-06,
95425
+ "loss": 0.4873,
95426
+ "step": 15714
95427
+ },
95428
+ {
95429
+ "epoch": 126.28,
95430
+ "learning_rate": 9.76237479806139e-06,
95431
+ "loss": 0.4662,
95432
+ "step": 15715
95433
+ },
95434
+ {
95435
+ "epoch": 126.29,
95436
+ "learning_rate": 9.762358642972537e-06,
95437
+ "loss": 0.5158,
95438
+ "step": 15716
95439
+ },
95440
+ {
95441
+ "epoch": 126.3,
95442
+ "learning_rate": 9.762342487883684e-06,
95443
+ "loss": 0.5165,
95444
+ "step": 15717
95445
+ },
95446
+ {
95447
+ "epoch": 126.31,
95448
+ "learning_rate": 9.762326332794831e-06,
95449
+ "loss": 0.5794,
95450
+ "step": 15718
95451
+ },
95452
+ {
95453
+ "epoch": 126.31,
95454
+ "learning_rate": 9.762310177705978e-06,
95455
+ "loss": 0.5821,
95456
+ "step": 15719
95457
+ },
95458
+ {
95459
+ "epoch": 126.32,
95460
+ "learning_rate": 9.762294022617125e-06,
95461
+ "loss": 0.572,
95462
+ "step": 15720
95463
+ },
95464
+ {
95465
+ "epoch": 126.33,
95466
+ "learning_rate": 9.762277867528273e-06,
95467
+ "loss": 0.6601,
95468
+ "step": 15721
95469
+ },
95470
+ {
95471
+ "epoch": 126.34,
95472
+ "learning_rate": 9.762261712439418e-06,
95473
+ "loss": 0.5877,
95474
+ "step": 15722
95475
+ },
95476
+ {
95477
+ "epoch": 126.35,
95478
+ "learning_rate": 9.762245557350565e-06,
95479
+ "loss": 0.8757,
95480
+ "step": 15723
95481
+ },
95482
+ {
95483
+ "epoch": 126.35,
95484
+ "learning_rate": 9.762229402261714e-06,
95485
+ "loss": 1.1488,
95486
+ "step": 15724
95487
+ },
95488
+ {
95489
+ "epoch": 126.36,
95490
+ "learning_rate": 9.76221324717286e-06,
95491
+ "loss": 1.217,
95492
+ "step": 15725
95493
+ },
95494
+ {
95495
+ "epoch": 126.37,
95496
+ "learning_rate": 9.762197092084007e-06,
95497
+ "loss": 1.277,
95498
+ "step": 15726
95499
+ },
95500
+ {
95501
+ "epoch": 126.38,
95502
+ "learning_rate": 9.762180936995154e-06,
95503
+ "loss": 1.0353,
95504
+ "step": 15727
95505
+ },
95506
+ {
95507
+ "epoch": 126.39,
95508
+ "learning_rate": 9.762164781906301e-06,
95509
+ "loss": 1.3811,
95510
+ "step": 15728
95511
+ },
95512
+ {
95513
+ "epoch": 126.39,
95514
+ "learning_rate": 9.762148626817448e-06,
95515
+ "loss": 1.6129,
95516
+ "step": 15729
95517
+ },
95518
+ {
95519
+ "epoch": 126.4,
95520
+ "learning_rate": 9.762132471728595e-06,
95521
+ "loss": 2.3937,
95522
+ "step": 15730
95523
+ },
95524
+ {
95525
+ "epoch": 126.41,
95526
+ "learning_rate": 9.762116316639743e-06,
95527
+ "loss": 1.0865,
95528
+ "step": 15731
95529
+ },
95530
+ {
95531
+ "epoch": 126.42,
95532
+ "learning_rate": 9.76210016155089e-06,
95533
+ "loss": 0.5552,
95534
+ "step": 15732
95535
+ },
95536
+ {
95537
+ "epoch": 126.43,
95538
+ "learning_rate": 9.762084006462035e-06,
95539
+ "loss": 0.5094,
95540
+ "step": 15733
95541
+ },
95542
+ {
95543
+ "epoch": 126.43,
95544
+ "learning_rate": 9.762067851373184e-06,
95545
+ "loss": 0.5008,
95546
+ "step": 15734
95547
+ },
95548
+ {
95549
+ "epoch": 126.44,
95550
+ "learning_rate": 9.76205169628433e-06,
95551
+ "loss": 0.6364,
95552
+ "step": 15735
95553
+ },
95554
+ {
95555
+ "epoch": 126.45,
95556
+ "learning_rate": 9.762035541195477e-06,
95557
+ "loss": 0.6163,
95558
+ "step": 15736
95559
+ },
95560
+ {
95561
+ "epoch": 126.46,
95562
+ "learning_rate": 9.762019386106624e-06,
95563
+ "loss": 0.4472,
95564
+ "step": 15737
95565
+ },
95566
+ {
95567
+ "epoch": 126.47,
95568
+ "learning_rate": 9.762003231017771e-06,
95569
+ "loss": 0.4391,
95570
+ "step": 15738
95571
+ },
95572
+ {
95573
+ "epoch": 126.47,
95574
+ "learning_rate": 9.761987075928918e-06,
95575
+ "loss": 0.6205,
95576
+ "step": 15739
95577
+ },
95578
+ {
95579
+ "epoch": 126.48,
95580
+ "learning_rate": 9.761970920840066e-06,
95581
+ "loss": 0.527,
95582
+ "step": 15740
95583
+ },
95584
+ {
95585
+ "epoch": 126.49,
95586
+ "learning_rate": 9.761954765751213e-06,
95587
+ "loss": 0.6126,
95588
+ "step": 15741
95589
+ },
95590
+ {
95591
+ "epoch": 126.5,
95592
+ "learning_rate": 9.76193861066236e-06,
95593
+ "loss": 0.5423,
95594
+ "step": 15742
95595
+ },
95596
+ {
95597
+ "epoch": 126.51,
95598
+ "learning_rate": 9.761922455573505e-06,
95599
+ "loss": 0.9217,
95600
+ "step": 15743
95601
+ },
95602
+ {
95603
+ "epoch": 126.51,
95604
+ "learning_rate": 9.761906300484654e-06,
95605
+ "loss": 0.6065,
95606
+ "step": 15744
95607
+ },
95608
+ {
95609
+ "epoch": 126.52,
95610
+ "learning_rate": 9.7618901453958e-06,
95611
+ "loss": 0.5341,
95612
+ "step": 15745
95613
+ },
95614
+ {
95615
+ "epoch": 126.53,
95616
+ "learning_rate": 9.761873990306947e-06,
95617
+ "loss": 0.5606,
95618
+ "step": 15746
95619
+ },
95620
+ {
95621
+ "epoch": 126.54,
95622
+ "learning_rate": 9.761857835218094e-06,
95623
+ "loss": 0.6874,
95624
+ "step": 15747
95625
+ },
95626
+ {
95627
+ "epoch": 126.55,
95628
+ "learning_rate": 9.761841680129241e-06,
95629
+ "loss": 0.7411,
95630
+ "step": 15748
95631
+ },
95632
+ {
95633
+ "epoch": 126.55,
95634
+ "learning_rate": 9.761825525040388e-06,
95635
+ "loss": 0.6518,
95636
+ "step": 15749
95637
+ },
95638
+ {
95639
+ "epoch": 126.56,
95640
+ "learning_rate": 9.761809369951536e-06,
95641
+ "loss": 0.8828,
95642
+ "step": 15750
95643
+ },
95644
+ {
95645
+ "epoch": 126.57,
95646
+ "learning_rate": 9.761793214862683e-06,
95647
+ "loss": 1.0841,
95648
+ "step": 15751
95649
+ },
95650
+ {
95651
+ "epoch": 126.58,
95652
+ "learning_rate": 9.76177705977383e-06,
95653
+ "loss": 1.4353,
95654
+ "step": 15752
95655
+ },
95656
+ {
95657
+ "epoch": 126.59,
95658
+ "learning_rate": 9.761760904684977e-06,
95659
+ "loss": 1.5527,
95660
+ "step": 15753
95661
+ },
95662
+ {
95663
+ "epoch": 126.59,
95664
+ "learning_rate": 9.761744749596124e-06,
95665
+ "loss": 1.9138,
95666
+ "step": 15754
95667
+ },
95668
+ {
95669
+ "epoch": 126.6,
95670
+ "learning_rate": 9.76172859450727e-06,
95671
+ "loss": 2.3029,
95672
+ "step": 15755
95673
+ },
95674
+ {
95675
+ "epoch": 126.61,
95676
+ "learning_rate": 9.761712439418417e-06,
95677
+ "loss": 0.5944,
95678
+ "step": 15756
95679
+ },
95680
+ {
95681
+ "epoch": 126.62,
95682
+ "learning_rate": 9.761696284329564e-06,
95683
+ "loss": 0.5497,
95684
+ "step": 15757
95685
+ },
95686
+ {
95687
+ "epoch": 126.63,
95688
+ "learning_rate": 9.761680129240711e-06,
95689
+ "loss": 0.4628,
95690
+ "step": 15758
95691
+ },
95692
+ {
95693
+ "epoch": 126.63,
95694
+ "learning_rate": 9.761663974151858e-06,
95695
+ "loss": 0.4982,
95696
+ "step": 15759
95697
+ },
95698
+ {
95699
+ "epoch": 126.64,
95700
+ "learning_rate": 9.761647819063006e-06,
95701
+ "loss": 0.686,
95702
+ "step": 15760
95703
+ },
95704
+ {
95705
+ "epoch": 126.65,
95706
+ "learning_rate": 9.761631663974153e-06,
95707
+ "loss": 0.8299,
95708
+ "step": 15761
95709
+ },
95710
+ {
95711
+ "epoch": 126.66,
95712
+ "learning_rate": 9.7616155088853e-06,
95713
+ "loss": 0.545,
95714
+ "step": 15762
95715
+ },
95716
+ {
95717
+ "epoch": 126.67,
95718
+ "learning_rate": 9.761599353796447e-06,
95719
+ "loss": 0.4436,
95720
+ "step": 15763
95721
+ },
95722
+ {
95723
+ "epoch": 126.67,
95724
+ "learning_rate": 9.761583198707594e-06,
95725
+ "loss": 0.5101,
95726
+ "step": 15764
95727
+ },
95728
+ {
95729
+ "epoch": 126.68,
95730
+ "learning_rate": 9.76156704361874e-06,
95731
+ "loss": 0.9322,
95732
+ "step": 15765
95733
+ },
95734
+ {
95735
+ "epoch": 126.69,
95736
+ "learning_rate": 9.761550888529889e-06,
95737
+ "loss": 0.5393,
95738
+ "step": 15766
95739
+ },
95740
+ {
95741
+ "epoch": 126.7,
95742
+ "learning_rate": 9.761534733441034e-06,
95743
+ "loss": 0.6011,
95744
+ "step": 15767
95745
+ },
95746
+ {
95747
+ "epoch": 126.71,
95748
+ "learning_rate": 9.761518578352181e-06,
95749
+ "loss": 0.5847,
95750
+ "step": 15768
95751
+ },
95752
+ {
95753
+ "epoch": 126.71,
95754
+ "learning_rate": 9.761502423263328e-06,
95755
+ "loss": 0.4552,
95756
+ "step": 15769
95757
+ },
95758
+ {
95759
+ "epoch": 126.72,
95760
+ "learning_rate": 9.761486268174476e-06,
95761
+ "loss": 0.689,
95762
+ "step": 15770
95763
+ },
95764
+ {
95765
+ "epoch": 126.73,
95766
+ "learning_rate": 9.761470113085623e-06,
95767
+ "loss": 1.0727,
95768
+ "step": 15771
95769
+ },
95770
+ {
95771
+ "epoch": 126.74,
95772
+ "learning_rate": 9.76145395799677e-06,
95773
+ "loss": 0.6094,
95774
+ "step": 15772
95775
+ },
95776
+ {
95777
+ "epoch": 126.75,
95778
+ "learning_rate": 9.761437802907917e-06,
95779
+ "loss": 0.6266,
95780
+ "step": 15773
95781
+ },
95782
+ {
95783
+ "epoch": 126.76,
95784
+ "learning_rate": 9.761421647819064e-06,
95785
+ "loss": 0.9351,
95786
+ "step": 15774
95787
+ },
95788
+ {
95789
+ "epoch": 126.76,
95790
+ "learning_rate": 9.76140549273021e-06,
95791
+ "loss": 0.7153,
95792
+ "step": 15775
95793
+ },
95794
+ {
95795
+ "epoch": 126.77,
95796
+ "learning_rate": 9.761389337641359e-06,
95797
+ "loss": 1.178,
95798
+ "step": 15776
95799
+ },
95800
+ {
95801
+ "epoch": 126.78,
95802
+ "learning_rate": 9.761373182552504e-06,
95803
+ "loss": 1.0686,
95804
+ "step": 15777
95805
+ },
95806
+ {
95807
+ "epoch": 126.79,
95808
+ "learning_rate": 9.761357027463651e-06,
95809
+ "loss": 1.2392,
95810
+ "step": 15778
95811
+ },
95812
+ {
95813
+ "epoch": 126.8,
95814
+ "learning_rate": 9.7613408723748e-06,
95815
+ "loss": 1.7776,
95816
+ "step": 15779
95817
+ },
95818
+ {
95819
+ "epoch": 126.8,
95820
+ "learning_rate": 9.761324717285946e-06,
95821
+ "loss": 2.6374,
95822
+ "step": 15780
95823
+ },
95824
+ {
95825
+ "epoch": 126.81,
95826
+ "learning_rate": 9.761308562197093e-06,
95827
+ "loss": 0.4673,
95828
+ "step": 15781
95829
+ },
95830
+ {
95831
+ "epoch": 126.82,
95832
+ "learning_rate": 9.76129240710824e-06,
95833
+ "loss": 1.0402,
95834
+ "step": 15782
95835
+ },
95836
+ {
95837
+ "epoch": 126.83,
95838
+ "learning_rate": 9.761276252019387e-06,
95839
+ "loss": 0.6464,
95840
+ "step": 15783
95841
+ },
95842
+ {
95843
+ "epoch": 126.84,
95844
+ "learning_rate": 9.761260096930534e-06,
95845
+ "loss": 0.5038,
95846
+ "step": 15784
95847
+ },
95848
+ {
95849
+ "epoch": 126.84,
95850
+ "learning_rate": 9.76124394184168e-06,
95851
+ "loss": 0.4841,
95852
+ "step": 15785
95853
+ },
95854
+ {
95855
+ "epoch": 126.85,
95856
+ "learning_rate": 9.761227786752829e-06,
95857
+ "loss": 0.583,
95858
+ "step": 15786
95859
+ },
95860
+ {
95861
+ "epoch": 126.86,
95862
+ "learning_rate": 9.761211631663974e-06,
95863
+ "loss": 0.4365,
95864
+ "step": 15787
95865
+ },
95866
+ {
95867
+ "epoch": 126.87,
95868
+ "learning_rate": 9.761195476575121e-06,
95869
+ "loss": 0.5512,
95870
+ "step": 15788
95871
+ },
95872
+ {
95873
+ "epoch": 126.88,
95874
+ "learning_rate": 9.76117932148627e-06,
95875
+ "loss": 0.632,
95876
+ "step": 15789
95877
+ },
95878
+ {
95879
+ "epoch": 126.88,
95880
+ "learning_rate": 9.761163166397416e-06,
95881
+ "loss": 0.4535,
95882
+ "step": 15790
95883
+ },
95884
+ {
95885
+ "epoch": 126.89,
95886
+ "learning_rate": 9.761147011308563e-06,
95887
+ "loss": 0.568,
95888
+ "step": 15791
95889
+ },
95890
+ {
95891
+ "epoch": 126.9,
95892
+ "learning_rate": 9.76113085621971e-06,
95893
+ "loss": 0.5274,
95894
+ "step": 15792
95895
+ },
95896
+ {
95897
+ "epoch": 126.91,
95898
+ "learning_rate": 9.761114701130857e-06,
95899
+ "loss": 0.7144,
95900
+ "step": 15793
95901
+ },
95902
+ {
95903
+ "epoch": 126.92,
95904
+ "learning_rate": 9.761098546042004e-06,
95905
+ "loss": 0.8883,
95906
+ "step": 15794
95907
+ },
95908
+ {
95909
+ "epoch": 126.92,
95910
+ "learning_rate": 9.761082390953152e-06,
95911
+ "loss": 0.6095,
95912
+ "step": 15795
95913
+ },
95914
+ {
95915
+ "epoch": 126.93,
95916
+ "learning_rate": 9.761066235864299e-06,
95917
+ "loss": 0.5981,
95918
+ "step": 15796
95919
+ },
95920
+ {
95921
+ "epoch": 126.94,
95922
+ "learning_rate": 9.761050080775446e-06,
95923
+ "loss": 0.6315,
95924
+ "step": 15797
95925
+ },
95926
+ {
95927
+ "epoch": 126.95,
95928
+ "learning_rate": 9.761033925686591e-06,
95929
+ "loss": 0.7411,
95930
+ "step": 15798
95931
+ },
95932
+ {
95933
+ "epoch": 126.96,
95934
+ "learning_rate": 9.76101777059774e-06,
95935
+ "loss": 0.7242,
95936
+ "step": 15799
95937
+ },
95938
+ {
95939
+ "epoch": 126.96,
95940
+ "learning_rate": 9.761001615508886e-06,
95941
+ "loss": 1.054,
95942
+ "step": 15800
95943
+ },
95944
+ {
95945
+ "epoch": 126.97,
95946
+ "learning_rate": 9.760985460420033e-06,
95947
+ "loss": 0.9365,
95948
+ "step": 15801
95949
+ },
95950
+ {
95951
+ "epoch": 126.98,
95952
+ "learning_rate": 9.76096930533118e-06,
95953
+ "loss": 1.5309,
95954
+ "step": 15802
95955
+ },
95956
+ {
95957
+ "epoch": 126.99,
95958
+ "learning_rate": 9.760953150242327e-06,
95959
+ "loss": 1.3608,
95960
+ "step": 15803
95961
+ },
95962
+ {
95963
+ "epoch": 127.0,
95964
+ "learning_rate": 9.760936995153474e-06,
95965
+ "loss": 1.9172,
95966
+ "step": 15804
95967
+ },
95968
+ {
95969
+ "epoch": 127.0,
95970
+ "eval_loss": 0.5659117698669434,
95971
+ "eval_runtime": 37.6401,
95972
+ "eval_samples_per_second": 22.237,
95973
+ "eval_steps_per_second": 0.717,
95974
+ "eval_wer": 0.25577642485146335,
95975
+ "step": 15804
95976
  }
95977
  ],
95978
  "max_steps": 620000,
95979
  "num_train_epochs": 5000,
95980
+ "total_flos": 4.444207979201224e+19,
95981
  "trial_name": null,
95982
  "trial_params": null
95983
  }
model-bin/finetune/base/{checkpoint-14810 → checkpoint-15804}/training_args.bin RENAMED
File without changes
model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/config.json RENAMED
File without changes
model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/optimizer.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7bb8e1f56462f3f72bc13099f261b35ee180692c7e5140807c0f34ea7a5c4207
3
  size 722165009
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0b0173e561dc647b2ccb918a3d9094c805bcdd4223814517e2e20335652b554
3
  size 722165009
model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/preprocessor_config.json RENAMED
File without changes
model-bin/finetune/base/{checkpoint-14810 → checkpoint-16425}/pytorch_model.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d25fe294170bc83af774c638907f09a1043ae4a3c2244887a10ae2bbb46c4ab8
3
  size 377909911
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5da972e4ce3390134f4abe8b670747d61997ae5fa33b2db44c6e08b07bc7da1
3
  size 377909911
model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/rng_state.pth RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:afc435ad298be0441a91b4ba2076f5509b3e6dbd613029605326c3dd6783b5ff
3
- size 14567
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f217c46c79e68803eb1ada396d3eca1ad30d847ec10e08a4446e75a540e1c8e5
3
+ size 14503
model-bin/finetune/base/{checkpoint-14810 → checkpoint-16425}/scaler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fb9e42f2f23f6da70c74b977a8ba5f32a04691fc1431afb402846d8cf68db549
3
  size 559
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:163f1fdbadb289dd6acf9aeffc01b4352e94b51a574ec3812e48b2fe65a23ce3
3
  size 559
model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/scheduler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3748a81ee940e4add3b77d69a66c82ab36139021c8cb2a3ead31c508b195f237
3
  size 623
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84f3b880ebabb9c057b8d75559f96743d9906e4ee669273e866df8b63eb9404d
3
  size 623
model-bin/finetune/base/{checkpoint-14810 → checkpoint-16425}/trainer_state.json RENAMED
The diff for this file is too large to render. See raw diff
 
model-bin/finetune/base/{checkpoint-15183 → checkpoint-16425}/training_args.bin RENAMED
File without changes
model-bin/finetune/base/log/1629515621.4037716/events.out.tfevents.1629515621.35c5ee475cfb.1049.43 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68bb2d168f3deeb851e044ad0b1d7a033b8e64e21d161761e7d6f1158c9412be
3
+ size 4194
model-bin/finetune/base/log/1629516257.3050275/events.out.tfevents.1629516257.35c5ee475cfb.1049.45 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:363f43c36a51fc8036e261a7a9d3c7727f789086468025d5b7730468060f533c
3
+ size 4194
model-bin/finetune/base/log/1629516898.1683943/events.out.tfevents.1629516898.35c5ee475cfb.1049.47 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c77a3d270bf8ca2b113ce86901c8870c90e6f19ab9140e26ed062af5f372ff6
3
+ size 4194
model-bin/finetune/base/log/1629517537.5125346/events.out.tfevents.1629517537.35c5ee475cfb.1049.49 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b1066a55321dcb4b6ffcc66fe4a5c478db44f36f35ffe6fe89722baa937ce23
3
+ size 4194
model-bin/finetune/base/log/1629518168.3557167/events.out.tfevents.1629518168.35c5ee475cfb.1049.51 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10e0cc0dcedf49e4835ddb9a1ae432b9c3c501c18146251adabc1f798ba79ca2
3
+ size 4194
model-bin/finetune/base/log/1629518794.4714847/events.out.tfevents.1629518794.35c5ee475cfb.1049.53 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52c1cfc34003551f47e50bcfff408289786af9b993549f6ccaa905e7e4899d45
3
+ size 4194
model-bin/finetune/base/log/1629519424.9971457/events.out.tfevents.1629519424.35c5ee475cfb.1049.55 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58f3ccfeeacd1cfcf025bdc56db4a42c63efbd2670114de2c6d3e17fcf809fec
3
+ size 4194
model-bin/finetune/base/log/1629520050.9530234/events.out.tfevents.1629520050.35c5ee475cfb.1049.57 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:238427021f6fd977451ebb350ef45ba0fa0c13e84816ed5dc355232ac02b5088
3
+ size 4194
model-bin/finetune/base/log/1629520692.8460991/events.out.tfevents.1629520692.35c5ee475cfb.1049.59 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ecfbeee863be34ec7dff41755f90e6a4f8a3690ec513d0019d66783f8ee4c477
3
+ size 4194
model-bin/finetune/base/log/1629521428.9057782/events.out.tfevents.1629521428.35c5ee475cfb.1049.61 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9bb970e31e51eb4ff7db72e8e765a8eb142abc071d71d010c6449c30f0acdcf
3
+ size 4194
model-bin/finetune/base/log/events.out.tfevents.1629515621.35c5ee475cfb.1049.42 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9c1e8039f02f5a7072f32bd0358763185c917e9e3ed21d1b085870489c3d322
3
+ size 24078
model-bin/finetune/base/log/events.out.tfevents.1629516257.35c5ee475cfb.1049.44 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cf3fc6cdee8107271cc8320d340ff4d1260445d4efdafddd6fa54e3405a0677
3
+ size 24078
model-bin/finetune/base/log/events.out.tfevents.1629516898.35c5ee475cfb.1049.46 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5099b73b0776c3435a735c01ed3abdd84792340e79803e2eda88f79f3f760e4c
3
+ size 24235
model-bin/finetune/base/log/events.out.tfevents.1629517537.35c5ee475cfb.1049.48 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b26c03337503aad1f66a88c660e3efedbab1310378e2812d62858998c2f47503
3
+ size 24078
model-bin/finetune/base/log/events.out.tfevents.1629518168.35c5ee475cfb.1049.50 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc70fb69cb6f545cdb9f507d5520e1cbd15f71fb8fb33187c757a988d430f808
3
+ size 24078
model-bin/finetune/base/log/events.out.tfevents.1629518794.35c5ee475cfb.1049.52 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56be4b10b9027e8a68c62c5f31a119eb3982c1e567125bd049a28e954287ff40
3
+ size 24078
model-bin/finetune/base/log/events.out.tfevents.1629519424.35c5ee475cfb.1049.54 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1afef0a0b7cbb65373ab2e173a8d36b1ca3ff3c73be6fb2b1f3c225487c34aa2
3
+ size 24078
model-bin/finetune/base/log/events.out.tfevents.1629520050.35c5ee475cfb.1049.56 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb6d0564ce24f623e0b6b1e9122e15ee2d5eff668dd8165f76c95d8e7efe0a37
3
+ size 24078
model-bin/finetune/base/log/events.out.tfevents.1629520692.35c5ee475cfb.1049.58 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2b99e5b438c7aa0668e5d850ce63d0d1f5f23508761d4baf044109a25b53bbc
3
+ size 24078
model-bin/finetune/base/log/events.out.tfevents.1629521428.35c5ee475cfb.1049.60 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85b1fd8a13feae797eb9ec3649d87c058c98a3465e7bb02f0646ead1ee5b0e1f
3
+ size 24373