Training in progress, step 10, checkpoint

last-checkpoint/adapter_config.json

@@ -23,8 +23,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "v_proj"
+    "v_proj",
+    "q_proj"
   ],
   "task_type": null,
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3817bdac47941aa18c809cca5ac9efbb6b46c4006fc14104768df724ee8242b
+oid sha256:3c4bcc45e6a8da2ae160f9bd65758048abe4faf10b4343a1aedbe1bf55f148e2
 size 62969640

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9f618664b5498348481f486e1963fddf3dc94b99759ca6beed2f98fca4dee3e
+oid sha256:1d05d9fe8f5a25b04f4025ef8dfde06c9cb8706f062556efa4dddd6d869f62cb
 size 21304084

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f73514fbd7715a5444d57d8a29c96393312f999f2b3d122ec2a6d7e76ffb689d
+oid sha256:911c06249d601e638f8189d761307ccdbacc556cb7c9096b657158b80e91a9fb
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:691263e55c7bb34b6e090bd89cf079f362100d7bd758be475bd3e7091c1a4ed6
+oid sha256:b4849a6ac0a1d895740f1ab4eba9d346b8d898008d0cfe93dd108cd928d7c63e
 size 1064

last-checkpoint/trainer_state.json

@@ -1,99 +1,35 @@
 {
-  "best_metric": 72.41379310344827,
-  "best_model_checkpoint": "../openai/whisper-large-v3-finetuned/checkpoint-20",
-  "epoch": 7.09,
-  "eval_steps": 20,
-  "global_step": 100,
+  "best_metric": 31.16279069767442,
+  "best_model_checkpoint": "../openai/whisper-large-v3-finetuned/checkpoint-10",
+  "epoch": 3.033333333333333,
+  "eval_steps": 10,
+  "global_step": 10,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.07,
-      "grad_norm": 1.2736600637435913,
+      "epoch": 3.033333333333333,
+      "grad_norm": 1.5397560596466064,
       "learning_rate": 0.0001,
-      "loss": 0.054,
-      "step": 20
+      "loss": 3.2643,
+      "step": 10
     },
     {
-      "epoch": 1.07,
-      "eval_loss": 2.1966075897216797,
-      "eval_runtime": 68.9209,
-      "eval_samples_per_second": 0.145,
-      "eval_steps_per_second": 0.044,
-      "eval_wer": 72.41379310344827,
-      "step": 20
-    },
-    {
-      "epoch": 3.01,
-      "grad_norm": 0.4058435261249542,
-      "learning_rate": 0.0001,
-      "loss": 0.0226,
-      "step": 40
-    },
-    {
-      "epoch": 3.01,
-      "eval_loss": 2.285231351852417,
-      "eval_runtime": 69.6442,
-      "eval_samples_per_second": 0.144,
-      "eval_steps_per_second": 0.043,
-      "eval_wer": 79.3103448275862,
-      "step": 40
-    },
-    {
-      "epoch": 4.08,
-      "grad_norm": 0.26837655901908875,
-      "learning_rate": 0.0001,
-      "loss": 0.0124,
-      "step": 60
-    },
-    {
-      "epoch": 4.08,
-      "eval_loss": 2.316169023513794,
-      "eval_runtime": 217.2614,
-      "eval_samples_per_second": 0.046,
-      "eval_steps_per_second": 0.014,
-      "eval_wer": 147.1264367816092,
-      "step": 60
-    },
-    {
-      "epoch": 6.02,
-      "grad_norm": 0.08057563006877899,
-      "learning_rate": 0.0001,
-      "loss": 0.0061,
-      "step": 80
-    },
-    {
-      "epoch": 6.02,
-      "eval_loss": 2.2900302410125732,
-      "eval_runtime": 68.7884,
-      "eval_samples_per_second": 0.145,
-      "eval_steps_per_second": 0.044,
-      "eval_wer": 74.71264367816092,
-      "step": 80
-    },
-    {
-      "epoch": 7.09,
-      "grad_norm": 0.15883919596672058,
-      "learning_rate": 0.0001,
-      "loss": 0.0039,
-      "step": 100
-    },
-    {
-      "epoch": 7.09,
-      "eval_loss": 2.3415772914886475,
-      "eval_runtime": 216.593,
-      "eval_samples_per_second": 0.046,
-      "eval_steps_per_second": 0.014,
-      "eval_wer": 141.3793103448276,
-      "step": 100
+      "epoch": 3.033333333333333,
+      "eval_loss": 3.1313788890838623,
+      "eval_runtime": 127.0315,
+      "eval_samples_per_second": 0.157,
+      "eval_steps_per_second": 0.024,
+      "eval_wer": 31.16279069767442,
+      "step": 10
     }
   ],
-  "logging_steps": 20,
-  "max_steps": 100,
+  "logging_steps": 10,
+  "max_steps": 30,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
-  "save_steps": 20,
+  "save_steps": 10,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {
@@ -101,13 +37,13 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 1.32542251794432e+18,
-  "train_batch_size": 4,
+  "total_flos": 2.3349412233216e+17,
+  "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d67fe5c97768c67bbca82cd2f92b09330635f100897633240c3099087275ae4
+oid sha256:73a2e914d430abee1d595139dd59f39cb44c03ff7a239558cb1fd89c77c3cef7
 size 5368