Training in progress, step 1800, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:532b3c8dd0fc48025330f131147e3bf655f8eb0ccc1b6f1be715ffcd37eeb0e6
+oid sha256:327212e84ba8a347d12bec59ece90867c94c9182c0b2908575049152721ce4a1
 size 16406680

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a018a28b363e2fe865407a2add098a16543edec7054f4d6b3acab628cd2db0b
+oid sha256:17bc83c01c7d996d4bae04714be783adfcc9079cc0fc45cb1321e249c2e52693
 size 6573615

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d482856761d4619002198ae9b8fc4dbddbf83ef7b336054705a7e2ace9dc37f5
+oid sha256:055ad8c7bb3354d2c4d9cad3fc3149048847f95be5ca999100b4b2b158d986d5
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9a24a7d5288f0f1fae49cca8a8fa22aa94481eb5acf767408b35e0d501a89b8
+oid sha256:0e7e8108fcdb3a0486aefb31155f0b6d11fb09d6229467e1e333e5c291d89d9a
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e95c77d1c57a34d51faaae9500eeda620e307df8870032eb6ea3441b2e2dfba
+oid sha256:104dbcce8836684729bcaeff6b52ebfb14715e797d21f0ca4a447252176b4567
 size 1465

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 1600,
-  "best_metric": 46.76727833990739,
-  "best_model_checkpoint": "./whisper-large-v3-turbo-swahili/checkpoint-1600",
-  "epoch": 2.8776978417266186,
+  "best_global_step": 1800,
+  "best_metric": 46.73297890584806,
+  "best_model_checkpoint": "./whisper-large-v3-turbo-swahili/checkpoint-1800",
+  "epoch": 3.237410071942446,
   "eval_steps": 200,
-  "global_step": 1600,
+  "global_step": 1800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -304,6 +304,43 @@
       "eval_steps_per_second": 0.176,
       "eval_wer": 46.76727833990739,
       "step": 1600
+    },
+    {
+      "epoch": 2.9676258992805753,
+      "grad_norm": 0.8634299635887146,
+      "learning_rate": 1.3311007957559682e-05,
+      "loss": 0.5152,
+      "step": 1650
+    },
+    {
+      "epoch": 3.0575539568345325,
+      "grad_norm": 0.8060579895973206,
+      "learning_rate": 1.3211538461538462e-05,
+      "loss": 0.4906,
+      "step": 1700
+    },
+    {
+      "epoch": 3.147482014388489,
+      "grad_norm": 0.9339563846588135,
+      "learning_rate": 1.3112068965517243e-05,
+      "loss": 0.492,
+      "step": 1750
+    },
+    {
+      "epoch": 3.237410071942446,
+      "grad_norm": 0.8844484090805054,
+      "learning_rate": 1.3012599469496021e-05,
+      "loss": 0.514,
+      "step": 1800
+    },
+    {
+      "epoch": 3.237410071942446,
+      "eval_loss": 0.7331941723823547,
+      "eval_runtime": 393.5575,
+      "eval_samples_per_second": 1.779,
+      "eval_steps_per_second": 0.178,
+      "eval_wer": 46.73297890584806,
+      "step": 1800
     }
   ],
   "logging_steps": 50,
@@ -323,7 +360,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 9.869390776369152e+19,
+  "total_flos": 1.1101007357411328e+20,
   "train_batch_size": 36,
   "trial_name": null,
   "trial_params": null