End of training

README.md

@@ -4,11 +4,24 @@ license: apache-2.0
 base_model: openai/whisper-medium
 tags:
 - generated_from_trainer
+datasets:
+- rbcurzon/ph_dialect_asr
 metrics:
 - wer
 model-index:
 - name: whisper-medium-test
-  results: []
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: rbcurzon/ph_dialect_asr all
+      type: rbcurzon/ph_dialect_asr
+      args: all
+    metrics:
+    - name: Wer
+      type: wer
+      value: 0.12251221112488153
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,7 +29,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # whisper-medium-test
 
-This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
+This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the rbcurzon/ph_dialect_asr all dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.2824
 - Wer: 0.1225

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 4.934979423868313,
+    "eval_loss": 0.2823709547519684,
+    "eval_runtime": 1581.903,
+    "eval_samples": 2619,
+    "eval_samples_per_second": 1.656,
+    "eval_steps_per_second": 0.414,
+    "eval_wer": 0.12251221112488153,
+    "total_flos": 4.894412894502912e+19,
+    "train_loss": 0.1704294042487939,
+    "train_runtime": 13123.9045,
+    "train_samples": 9717,
+    "train_samples_per_second": 3.657,
+    "train_steps_per_second": 0.229
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.934979423868313,
+    "eval_loss": 0.2823709547519684,
+    "eval_runtime": 1581.903,
+    "eval_samples": 2619,
+    "eval_samples_per_second": 1.656,
+    "eval_steps_per_second": 0.414,
+    "eval_wer": 0.12251221112488153
+}

runs/Jul25_07-15-42_b0effe23cf42/events.out.tfevents.1753443510.b0effe23cf42.3246.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42690f57208cc5c1bbf698779758a040e1b009ea7e9521c5bc9c1f7727e9c4b4
+size 406

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.934979423868313,
+    "total_flos": 4.894412894502912e+19,
+    "train_loss": 0.1704294042487939,
+    "train_runtime": 13123.9045,
+    "train_samples": 9717,
+    "train_samples_per_second": 3.657,
+    "train_steps_per_second": 0.229
+}

trainer_state.json

@@ -0,0 +1,910 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.934979423868313,
+  "eval_steps": 1000,
+  "global_step": 3000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0411522633744856,
+      "grad_norm": 14.940086364746094,
+      "learning_rate": 4.6000000000000004e-07,
+      "loss": 1.357,
+      "step": 25
+    },
+    {
+      "epoch": 0.0823045267489712,
+      "grad_norm": 9.376535415649414,
+      "learning_rate": 9.600000000000001e-07,
+      "loss": 1.124,
+      "step": 50
+    },
+    {
+      "epoch": 0.12345679012345678,
+      "grad_norm": 8.351006507873535,
+      "learning_rate": 1.44e-06,
+      "loss": 0.9374,
+      "step": 75
+    },
+    {
+      "epoch": 0.1646090534979424,
+      "grad_norm": 9.542637825012207,
+      "learning_rate": 1.94e-06,
+      "loss": 0.7622,
+      "step": 100
+    },
+    {
+      "epoch": 0.205761316872428,
+      "grad_norm": 9.52396297454834,
+      "learning_rate": 2.4400000000000004e-06,
+      "loss": 0.7141,
+      "step": 125
+    },
+    {
+      "epoch": 0.24691358024691357,
+      "grad_norm": 7.540445804595947,
+      "learning_rate": 2.9400000000000002e-06,
+      "loss": 0.6652,
+      "step": 150
+    },
+    {
+      "epoch": 0.2880658436213992,
+      "grad_norm": 7.401683807373047,
+      "learning_rate": 3.44e-06,
+      "loss": 0.5939,
+      "step": 175
+    },
+    {
+      "epoch": 0.3292181069958848,
+      "grad_norm": 9.240742683410645,
+      "learning_rate": 3.94e-06,
+      "loss": 0.5542,
+      "step": 200
+    },
+    {
+      "epoch": 0.37037037037037035,
+      "grad_norm": 7.216451644897461,
+      "learning_rate": 4.440000000000001e-06,
+      "loss": 0.544,
+      "step": 225
+    },
+    {
+      "epoch": 0.411522633744856,
+      "grad_norm": 7.483836650848389,
+      "learning_rate": 4.94e-06,
+      "loss": 0.5081,
+      "step": 250
+    },
+    {
+      "epoch": 0.45267489711934156,
+      "grad_norm": 6.959393501281738,
+      "learning_rate": 5.4400000000000004e-06,
+      "loss": 0.4608,
+      "step": 275
+    },
+    {
+      "epoch": 0.49382716049382713,
+      "grad_norm": 5.851141452789307,
+      "learning_rate": 5.94e-06,
+      "loss": 0.4661,
+      "step": 300
+    },
+    {
+      "epoch": 0.5349794238683128,
+      "grad_norm": 6.5286712646484375,
+      "learning_rate": 6.440000000000001e-06,
+      "loss": 0.4301,
+      "step": 325
+    },
+    {
+      "epoch": 0.5761316872427984,
+      "grad_norm": 7.329245567321777,
+      "learning_rate": 6.9400000000000005e-06,
+      "loss": 0.4293,
+      "step": 350
+    },
+    {
+      "epoch": 0.6172839506172839,
+      "grad_norm": 7.156967639923096,
+      "learning_rate": 7.440000000000001e-06,
+      "loss": 0.4247,
+      "step": 375
+    },
+    {
+      "epoch": 0.6584362139917695,
+      "grad_norm": 6.780092716217041,
+      "learning_rate": 7.94e-06,
+      "loss": 0.4084,
+      "step": 400
+    },
+    {
+      "epoch": 0.6995884773662552,
+      "grad_norm": 5.79763126373291,
+      "learning_rate": 8.44e-06,
+      "loss": 0.3924,
+      "step": 425
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 6.162405014038086,
+      "learning_rate": 8.94e-06,
+      "loss": 0.3641,
+      "step": 450
+    },
+    {
+      "epoch": 0.7818930041152263,
+      "grad_norm": 6.669307231903076,
+      "learning_rate": 9.440000000000001e-06,
+      "loss": 0.3631,
+      "step": 475
+    },
+    {
+      "epoch": 0.823045267489712,
+      "grad_norm": 5.827241897583008,
+      "learning_rate": 9.940000000000001e-06,
+      "loss": 0.3607,
+      "step": 500
+    },
+    {
+      "epoch": 0.8641975308641975,
+      "grad_norm": 7.120790004730225,
+      "learning_rate": 9.912000000000001e-06,
+      "loss": 0.3396,
+      "step": 525
+    },
+    {
+      "epoch": 0.9053497942386831,
+      "grad_norm": 5.817197322845459,
+      "learning_rate": 9.812e-06,
+      "loss": 0.3416,
+      "step": 550
+    },
+    {
+      "epoch": 0.9465020576131687,
+      "grad_norm": 6.041031837463379,
+      "learning_rate": 9.712e-06,
+      "loss": 0.3415,
+      "step": 575
+    },
+    {
+      "epoch": 0.9876543209876543,
+      "grad_norm": 5.69822883605957,
+      "learning_rate": 9.612000000000002e-06,
+      "loss": 0.3304,
+      "step": 600
+    },
+    {
+      "epoch": 1.0279835390946501,
+      "grad_norm": 5.423857688903809,
+      "learning_rate": 9.512000000000001e-06,
+      "loss": 0.2518,
+      "step": 625
+    },
+    {
+      "epoch": 1.0691358024691358,
+      "grad_norm": 4.724055767059326,
+      "learning_rate": 9.412e-06,
+      "loss": 0.2157,
+      "step": 650
+    },
+    {
+      "epoch": 1.1102880658436214,
+      "grad_norm": 4.128566265106201,
+      "learning_rate": 9.312000000000002e-06,
+      "loss": 0.2118,
+      "step": 675
+    },
+    {
+      "epoch": 1.151440329218107,
+      "grad_norm": 4.85585880279541,
+      "learning_rate": 9.212000000000001e-06,
+      "loss": 0.2327,
+      "step": 700
+    },
+    {
+      "epoch": 1.1925925925925926,
+      "grad_norm": 5.799758434295654,
+      "learning_rate": 9.112e-06,
+      "loss": 0.2041,
+      "step": 725
+    },
+    {
+      "epoch": 1.2337448559670783,
+      "grad_norm": 3.91048002243042,
+      "learning_rate": 9.012000000000001e-06,
+      "loss": 0.1966,
+      "step": 750
+    },
+    {
+      "epoch": 1.274897119341564,
+      "grad_norm": 5.111474990844727,
+      "learning_rate": 8.912e-06,
+      "loss": 0.2009,
+      "step": 775
+    },
+    {
+      "epoch": 1.3160493827160493,
+      "grad_norm": 5.129931449890137,
+      "learning_rate": 8.812000000000001e-06,
+      "loss": 0.177,
+      "step": 800
+    },
+    {
+      "epoch": 1.357201646090535,
+      "grad_norm": 3.5538148880004883,
+      "learning_rate": 8.712e-06,
+      "loss": 0.1988,
+      "step": 825
+    },
+    {
+      "epoch": 1.3983539094650206,
+      "grad_norm": 4.522074222564697,
+      "learning_rate": 8.612e-06,
+      "loss": 0.1998,
+      "step": 850
+    },
+    {
+      "epoch": 1.4395061728395062,
+      "grad_norm": 4.1635637283325195,
+      "learning_rate": 8.512e-06,
+      "loss": 0.1815,
+      "step": 875
+    },
+    {
+      "epoch": 1.4806584362139918,
+      "grad_norm": 4.162914752960205,
+      "learning_rate": 8.412e-06,
+      "loss": 0.181,
+      "step": 900
+    },
+    {
+      "epoch": 1.5218106995884773,
+      "grad_norm": 5.481962203979492,
+      "learning_rate": 8.312000000000001e-06,
+      "loss": 0.1817,
+      "step": 925
+    },
+    {
+      "epoch": 1.5629629629629629,
+      "grad_norm": 4.371176242828369,
+      "learning_rate": 8.212e-06,
+      "loss": 0.1855,
+      "step": 950
+    },
+    {
+      "epoch": 1.6041152263374485,
+      "grad_norm": 4.259627819061279,
+      "learning_rate": 8.112000000000001e-06,
+      "loss": 0.1768,
+      "step": 975
+    },
+    {
+      "epoch": 1.6452674897119342,
+      "grad_norm": 4.984127044677734,
+      "learning_rate": 8.012e-06,
+      "loss": 0.1981,
+      "step": 1000
+    },
+    {
+      "epoch": 1.6452674897119342,
+      "eval_loss": 0.2714756429195404,
+      "eval_runtime": 1557.3098,
+      "eval_samples_per_second": 1.682,
+      "eval_steps_per_second": 0.421,
+      "eval_wer": 0.14547641612597506,
+      "step": 1000
+    },
+    {
+      "epoch": 1.6864197530864198,
+      "grad_norm": 3.931295156478882,
+      "learning_rate": 7.912000000000001e-06,
+      "loss": 0.1774,
+      "step": 1025
+    },
+    {
+      "epoch": 1.7275720164609054,
+      "grad_norm": 4.057419300079346,
+      "learning_rate": 7.812e-06,
+      "loss": 0.1789,
+      "step": 1050
+    },
+    {
+      "epoch": 1.768724279835391,
+      "grad_norm": 3.723196029663086,
+      "learning_rate": 7.712e-06,
+      "loss": 0.1703,
+      "step": 1075
+    },
+    {
+      "epoch": 1.8098765432098767,
+      "grad_norm": 5.39990758895874,
+      "learning_rate": 7.612e-06,
+      "loss": 0.1815,
+      "step": 1100
+    },
+    {
+      "epoch": 1.851028806584362,
+      "grad_norm": 4.207727909088135,
+      "learning_rate": 7.512e-06,
+      "loss": 0.1561,
+      "step": 1125
+    },
+    {
+      "epoch": 1.8921810699588477,
+      "grad_norm": 5.286857604980469,
+      "learning_rate": 7.412e-06,
+      "loss": 0.1848,
+      "step": 1150
+    },
+    {
+      "epoch": 1.9333333333333333,
+      "grad_norm": 3.916595697402954,
+      "learning_rate": 7.3120000000000005e-06,
+      "loss": 0.1673,
+      "step": 1175
+    },
+    {
+      "epoch": 1.974485596707819,
+      "grad_norm": 5.331394195556641,
+      "learning_rate": 7.212e-06,
+      "loss": 0.1939,
+      "step": 1200
+    },
+    {
+      "epoch": 2.0148148148148146,
+      "grad_norm": 2.4200124740600586,
+      "learning_rate": 7.1120000000000015e-06,
+      "loss": 0.1235,
+      "step": 1225
+    },
+    {
+      "epoch": 2.0559670781893002,
+      "grad_norm": 6.067932605743408,
+      "learning_rate": 7.012000000000001e-06,
+      "loss": 0.0826,
+      "step": 1250
+    },
+    {
+      "epoch": 2.097119341563786,
+      "grad_norm": 3.729630708694458,
+      "learning_rate": 6.912000000000001e-06,
+      "loss": 0.0721,
+      "step": 1275
+    },
+    {
+      "epoch": 2.1382716049382715,
+      "grad_norm": 2.2330338954925537,
+      "learning_rate": 6.812000000000001e-06,
+      "loss": 0.0712,
+      "step": 1300
+    },
+    {
+      "epoch": 2.179423868312757,
+      "grad_norm": 2.977776288986206,
+      "learning_rate": 6.712000000000001e-06,
+      "loss": 0.069,
+      "step": 1325
+    },
+    {
+      "epoch": 2.2205761316872428,
+      "grad_norm": 3.0866611003875732,
+      "learning_rate": 6.612e-06,
+      "loss": 0.077,
+      "step": 1350
+    },
+    {
+      "epoch": 2.2617283950617284,
+      "grad_norm": 1.9583996534347534,
+      "learning_rate": 6.5120000000000005e-06,
+      "loss": 0.0681,
+      "step": 1375
+    },
+    {
+      "epoch": 2.302880658436214,
+      "grad_norm": 3.112598419189453,
+      "learning_rate": 6.412000000000001e-06,
+      "loss": 0.0766,
+      "step": 1400
+    },
+    {
+      "epoch": 2.3440329218106997,
+      "grad_norm": 3.2026150226593018,
+      "learning_rate": 6.312000000000001e-06,
+      "loss": 0.0776,
+      "step": 1425
+    },
+    {
+      "epoch": 2.3851851851851853,
+      "grad_norm": 3.7830779552459717,
+      "learning_rate": 6.212e-06,
+      "loss": 0.0763,
+      "step": 1450
+    },
+    {
+      "epoch": 2.426337448559671,
+      "grad_norm": 1.5785398483276367,
+      "learning_rate": 6.112e-06,
+      "loss": 0.0666,
+      "step": 1475
+    },
+    {
+      "epoch": 2.4674897119341566,
+      "grad_norm": 3.071021795272827,
+      "learning_rate": 6.012e-06,
+      "loss": 0.074,
+      "step": 1500
+    },
+    {
+      "epoch": 2.5086419753086417,
+      "grad_norm": 2.2543785572052,
+      "learning_rate": 5.912e-06,
+      "loss": 0.0711,
+      "step": 1525
+    },
+    {
+      "epoch": 2.549794238683128,
+      "grad_norm": 2.8244235515594482,
+      "learning_rate": 5.812000000000001e-06,
+      "loss": 0.068,
+      "step": 1550
+    },
+    {
+      "epoch": 2.590946502057613,
+      "grad_norm": 1.933445930480957,
+      "learning_rate": 5.7120000000000005e-06,
+      "loss": 0.0797,
+      "step": 1575
+    },
+    {
+      "epoch": 2.6320987654320986,
+      "grad_norm": 2.2213869094848633,
+      "learning_rate": 5.612000000000001e-06,
+      "loss": 0.072,
+      "step": 1600
+    },
+    {
+      "epoch": 2.6732510288065843,
+      "grad_norm": 3.0998716354370117,
+      "learning_rate": 5.512000000000001e-06,
+      "loss": 0.0638,
+      "step": 1625
+    },
+    {
+      "epoch": 2.71440329218107,
+      "grad_norm": 3.967010974884033,
+      "learning_rate": 5.412000000000001e-06,
+      "loss": 0.0728,
+      "step": 1650
+    },
+    {
+      "epoch": 2.7555555555555555,
+      "grad_norm": 1.7496695518493652,
+      "learning_rate": 5.312e-06,
+      "loss": 0.0677,
+      "step": 1675
+    },
+    {
+      "epoch": 2.796707818930041,
+      "grad_norm": 3.4264888763427734,
+      "learning_rate": 5.212e-06,
+      "loss": 0.0711,
+      "step": 1700
+    },
+    {
+      "epoch": 2.837860082304527,
+      "grad_norm": 3.8687572479248047,
+      "learning_rate": 5.112e-06,
+      "loss": 0.0694,
+      "step": 1725
+    },
+    {
+      "epoch": 2.8790123456790124,
+      "grad_norm": 2.948435068130493,
+      "learning_rate": 5.0120000000000005e-06,
+      "loss": 0.0734,
+      "step": 1750
+    },
+    {
+      "epoch": 2.920164609053498,
+      "grad_norm": 2.5124669075012207,
+      "learning_rate": 4.9120000000000006e-06,
+      "loss": 0.0789,
+      "step": 1775
+    },
+    {
+      "epoch": 2.9613168724279837,
+      "grad_norm": 2.759510040283203,
+      "learning_rate": 4.812000000000001e-06,
+      "loss": 0.0616,
+      "step": 1800
+    },
+    {
+      "epoch": 3.0016460905349795,
+      "grad_norm": 1.3032381534576416,
+      "learning_rate": 4.712000000000001e-06,
+      "loss": 0.0516,
+      "step": 1825
+    },
+    {
+      "epoch": 3.042798353909465,
+      "grad_norm": 1.752594232559204,
+      "learning_rate": 4.612e-06,
+      "loss": 0.0232,
+      "step": 1850
+    },
+    {
+      "epoch": 3.083950617283951,
+      "grad_norm": 1.5227144956588745,
+      "learning_rate": 4.512e-06,
+      "loss": 0.0295,
+      "step": 1875
+    },
+    {
+      "epoch": 3.125102880658436,
+      "grad_norm": 1.635178565979004,
+      "learning_rate": 4.412e-06,
+      "loss": 0.0244,
+      "step": 1900
+    },
+    {
+      "epoch": 3.1662551440329216,
+      "grad_norm": 1.666857123374939,
+      "learning_rate": 4.312e-06,
+      "loss": 0.0232,
+      "step": 1925
+    },
+    {
+      "epoch": 3.2074074074074073,
+      "grad_norm": 1.3419064283370972,
+      "learning_rate": 4.2120000000000005e-06,
+      "loss": 0.0245,
+      "step": 1950
+    },
+    {
+      "epoch": 3.248559670781893,
+      "grad_norm": 1.2259336709976196,
+      "learning_rate": 4.112000000000001e-06,
+      "loss": 0.0286,
+      "step": 1975
+    },
+    {
+      "epoch": 3.2897119341563785,
+      "grad_norm": 1.062391757965088,
+      "learning_rate": 4.012000000000001e-06,
+      "loss": 0.0236,
+      "step": 2000
+    },
+    {
+      "epoch": 3.2897119341563785,
+      "eval_loss": 0.27091845870018005,
+      "eval_runtime": 1588.6268,
+      "eval_samples_per_second": 1.649,
+      "eval_steps_per_second": 0.412,
+      "eval_wer": 0.12672231537508202,
+      "step": 2000
+    },
+    {
+      "epoch": 3.330864197530864,
+      "grad_norm": 2.021160840988159,
+      "learning_rate": 3.912e-06,
+      "loss": 0.0256,
+      "step": 2025
+    },
+    {
+      "epoch": 3.37201646090535,
+      "grad_norm": 1.8684940338134766,
+      "learning_rate": 3.812e-06,
+      "loss": 0.026,
+      "step": 2050
+    },
+    {
+      "epoch": 3.4131687242798354,
+      "grad_norm": 0.7061536908149719,
+      "learning_rate": 3.712e-06,
+      "loss": 0.0225,
+      "step": 2075
+    },
+    {
+      "epoch": 3.454320987654321,
+      "grad_norm": 1.89629328250885,
+      "learning_rate": 3.6120000000000003e-06,
+      "loss": 0.0273,
+      "step": 2100
+    },
+    {
+      "epoch": 3.4954732510288067,
+      "grad_norm": 1.4353647232055664,
+      "learning_rate": 3.5120000000000004e-06,
+      "loss": 0.0231,
+      "step": 2125
+    },
+    {
+      "epoch": 3.5366255144032923,
+      "grad_norm": 0.97385573387146,
+      "learning_rate": 3.412e-06,
+      "loss": 0.0224,
+      "step": 2150
+    },
+    {
+      "epoch": 3.5777777777777775,
+      "grad_norm": 1.3879934549331665,
+      "learning_rate": 3.3120000000000002e-06,
+      "loss": 0.0247,
+      "step": 2175
+    },
+    {
+      "epoch": 3.6189300411522636,
+      "grad_norm": 1.5851539373397827,
+      "learning_rate": 3.212e-06,
+      "loss": 0.0239,
+      "step": 2200
+    },
+    {
+      "epoch": 3.6600823045267488,
+      "grad_norm": 0.3798752725124359,
+      "learning_rate": 3.112e-06,
+      "loss": 0.0268,
+      "step": 2225
+    },
+    {
+      "epoch": 3.701234567901235,
+      "grad_norm": 2.217914342880249,
+      "learning_rate": 3.0120000000000006e-06,
+      "loss": 0.0226,
+      "step": 2250
+    },
+    {
+      "epoch": 3.74238683127572,
+      "grad_norm": 1.0964093208312988,
+      "learning_rate": 2.9120000000000002e-06,
+      "loss": 0.0345,
+      "step": 2275
+    },
+    {
+      "epoch": 3.7835390946502057,
+      "grad_norm": 0.7668254375457764,
+      "learning_rate": 2.8120000000000004e-06,
+      "loss": 0.0247,
+      "step": 2300
+    },
+    {
+      "epoch": 3.8246913580246913,
+      "grad_norm": 2.1343576908111572,
+      "learning_rate": 2.712e-06,
+      "loss": 0.0188,
+      "step": 2325
+    },
+    {
+      "epoch": 3.865843621399177,
+      "grad_norm": 3.1728477478027344,
+      "learning_rate": 2.612e-06,
+      "loss": 0.0304,
+      "step": 2350
+    },
+    {
+      "epoch": 3.9069958847736626,
+      "grad_norm": 0.8707403540611267,
+      "learning_rate": 2.512e-06,
+      "loss": 0.0287,
+      "step": 2375
+    },
+    {
+      "epoch": 3.948148148148148,
+      "grad_norm": 1.3772474527359009,
+      "learning_rate": 2.4120000000000004e-06,
+      "loss": 0.021,
+      "step": 2400
+    },
+    {
+      "epoch": 3.989300411522634,
+      "grad_norm": 0.8075869083404541,
+      "learning_rate": 2.312e-06,
+      "loss": 0.0224,
+      "step": 2425
+    },
+    {
+      "epoch": 4.029629629629629,
+      "grad_norm": 0.8268884420394897,
+      "learning_rate": 2.212e-06,
+      "loss": 0.0114,
+      "step": 2450
+    },
+    {
+      "epoch": 4.070781893004115,
+      "grad_norm": 0.6684131026268005,
+      "learning_rate": 2.1120000000000003e-06,
+      "loss": 0.008,
+      "step": 2475
+    },
+    {
+      "epoch": 4.1119341563786005,
+      "grad_norm": 0.2903365194797516,
+      "learning_rate": 2.012e-06,
+      "loss": 0.0148,
+      "step": 2500
+    },
+    {
+      "epoch": 4.153086419753087,
+      "grad_norm": 2.815356731414795,
+      "learning_rate": 1.912e-06,
+      "loss": 0.0112,
+      "step": 2525
+    },
+    {
+      "epoch": 4.194238683127572,
+      "grad_norm": 0.7152761816978455,
+      "learning_rate": 1.8120000000000002e-06,
+      "loss": 0.0135,
+      "step": 2550
+    },
+    {
+      "epoch": 4.235390946502058,
+      "grad_norm": 0.7304953932762146,
+      "learning_rate": 1.712e-06,
+      "loss": 0.0076,
+      "step": 2575
+    },
+    {
+      "epoch": 4.276543209876543,
+      "grad_norm": 1.0283336639404297,
+      "learning_rate": 1.6120000000000002e-06,
+      "loss": 0.0138,
+      "step": 2600
+    },
+    {
+      "epoch": 4.317695473251029,
+      "grad_norm": 0.19157454371452332,
+      "learning_rate": 1.512e-06,
+      "loss": 0.0059,
+      "step": 2625
+    },
+    {
+      "epoch": 4.358847736625514,
+      "grad_norm": 0.8817554116249084,
+      "learning_rate": 1.412e-06,
+      "loss": 0.0059,
+      "step": 2650
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 0.772783637046814,
+      "learning_rate": 1.3120000000000003e-06,
+      "loss": 0.0067,
+      "step": 2675
+    },
+    {
+      "epoch": 4.4411522633744855,
+      "grad_norm": 1.7318344116210938,
+      "learning_rate": 1.2120000000000002e-06,
+      "loss": 0.0091,
+      "step": 2700
+    },
+    {
+      "epoch": 4.482304526748971,
+      "grad_norm": 0.4040825366973877,
+      "learning_rate": 1.1120000000000001e-06,
+      "loss": 0.0075,
+      "step": 2725
+    },
+    {
+      "epoch": 4.523456790123457,
+      "grad_norm": 0.22325092554092407,
+      "learning_rate": 1.012e-06,
+      "loss": 0.0068,
+      "step": 2750
+    },
+    {
+      "epoch": 4.564609053497943,
+      "grad_norm": 0.44631338119506836,
+      "learning_rate": 9.120000000000001e-07,
+      "loss": 0.0065,
+      "step": 2775
+    },
+    {
+      "epoch": 4.605761316872428,
+      "grad_norm": 1.0319997072219849,
+      "learning_rate": 8.12e-07,
+      "loss": 0.0071,
+      "step": 2800
+    },
+    {
+      "epoch": 4.646913580246913,
+      "grad_norm": 1.9721506834030151,
+      "learning_rate": 7.12e-07,
+      "loss": 0.0119,
+      "step": 2825
+    },
+    {
+      "epoch": 4.688065843621399,
+      "grad_norm": 0.18514037132263184,
+      "learning_rate": 6.12e-07,
+      "loss": 0.0067,
+      "step": 2850
+    },
+    {
+      "epoch": 4.7292181069958845,
+      "grad_norm": 0.5566908121109009,
+      "learning_rate": 5.12e-07,
+      "loss": 0.0077,
+      "step": 2875
+    },
+    {
+      "epoch": 4.770370370370371,
+      "grad_norm": 0.2684698700904846,
+      "learning_rate": 4.1200000000000004e-07,
+      "loss": 0.0093,
+      "step": 2900
+    },
+    {
+      "epoch": 4.811522633744856,
+      "grad_norm": 0.36519676446914673,
+      "learning_rate": 3.12e-07,
+      "loss": 0.0077,
+      "step": 2925
+    },
+    {
+      "epoch": 4.852674897119342,
+      "grad_norm": 0.5899785161018372,
+      "learning_rate": 2.1200000000000002e-07,
+      "loss": 0.0059,
+      "step": 2950
+    },
+    {
+      "epoch": 4.893827160493827,
+      "grad_norm": 0.5213066339492798,
+      "learning_rate": 1.1200000000000001e-07,
+      "loss": 0.0051,
+      "step": 2975
+    },
+    {
+      "epoch": 4.934979423868313,
+      "grad_norm": 0.2481735646724701,
+      "learning_rate": 1.2e-08,
+      "loss": 0.0066,
+      "step": 3000
+    },
+    {
+      "epoch": 4.934979423868313,
+      "eval_loss": 0.2823709547519684,
+      "eval_runtime": 1578.0998,
+      "eval_samples_per_second": 1.66,
+      "eval_steps_per_second": 0.415,
+      "eval_wer": 0.12251221112488153,
+      "step": 3000
+    },
+    {
+      "epoch": 4.934979423868313,
+      "step": 3000,
+      "total_flos": 4.894412894502912e+19,
+      "train_loss": 0.1704294042487939,
+      "train_runtime": 13123.9045,
+      "train_samples_per_second": 3.657,
+      "train_steps_per_second": 0.229
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.894412894502912e+19,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}