End of training

README.md

@@ -3,6 +3,7 @@ library_name: transformers
 license: apache-2.0
 base_model: facebook/wav2vec2-base
 tags:
+- audio-classification
 - generated_from_trainer
 metrics:
 - accuracy
@@ -16,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-base-lang-id
 
-This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
+This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the /content/dataset dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0127
+- Loss: 0.0134
 - Accuracy: 0.998
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 4.992506244796004,
+    "eval_accuracy": 0.998,
+    "eval_loss": 0.013417072594165802,
+    "eval_runtime": 42.1226,
+    "eval_samples_per_second": 47.48,
+    "eval_steps_per_second": 11.87,
+    "total_flos": 1.2141162127800926e+19,
+    "train_loss": 0.10829704528961641,
+    "train_runtime": 6533.0547,
+    "train_samples_per_second": 22.052,
+    "train_steps_per_second": 0.459
+}

runs/Jan15_09-44-36_6da9dd5ea643/events.out.tfevents.1736940869.6da9dd5ea643.11685.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb16706c1b42002e0d35eeeb5448ade9dd9a36a63f9118a709bf2b0d83ef6a4e
+size 411

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.992506244796004,
+    "eval_accuracy": 0.998,
+    "eval_loss": 0.013417072594165802,
+    "eval_runtime": 42.1226,
+    "eval_samples_per_second": 47.48,
+    "eval_steps_per_second": 11.87
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.992506244796004,
+    "total_flos": 1.2141162127800926e+19,
+    "train_loss": 0.10829704528961641,
+    "train_runtime": 6533.0547,
+    "train_samples_per_second": 22.052,
+    "train_steps_per_second": 0.459
+}

trainer_state.json

@@ -0,0 +1,2187 @@
+{
+  "best_metric": 0.998,
+  "best_model_checkpoint": "wav2vec2-base-lang-id/checkpoint-2404",
+  "epoch": 4.992506244796004,
+  "eval_steps": 500,
+  "global_step": 3000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01665278934221482,
+      "grad_norm": 0.7952147126197815,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 2.1922,
+      "step": 10
+    },
+    {
+      "epoch": 0.03330557868442964,
+      "grad_norm": 0.8025399446487427,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 2.1899,
+      "step": 20
+    },
+    {
+      "epoch": 0.04995836802664446,
+      "grad_norm": 1.0220482349395752,
+      "learning_rate": 6e-06,
+      "loss": 2.1615,
+      "step": 30
+    },
+    {
+      "epoch": 0.06661115736885928,
+      "grad_norm": 1.5168348550796509,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 2.1031,
+      "step": 40
+    },
+    {
+      "epoch": 0.08326394671107411,
+      "grad_norm": 2.3170907497406006,
+      "learning_rate": 1e-05,
+      "loss": 1.9853,
+      "step": 50
+    },
+    {
+      "epoch": 0.09991673605328892,
+      "grad_norm": 2.646939992904663,
+      "learning_rate": 1.2e-05,
+      "loss": 1.819,
+      "step": 60
+    },
+    {
+      "epoch": 0.11656952539550375,
+      "grad_norm": 3.0145034790039062,
+      "learning_rate": 1.4e-05,
+      "loss": 1.5971,
+      "step": 70
+    },
+    {
+      "epoch": 0.13322231473771856,
+      "grad_norm": 6.5615739822387695,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 1.3221,
+      "step": 80
+    },
+    {
+      "epoch": 0.1498751040799334,
+      "grad_norm": 7.5968146324157715,
+      "learning_rate": 1.8e-05,
+      "loss": 1.0419,
+      "step": 90
+    },
+    {
+      "epoch": 0.16652789342214822,
+      "grad_norm": 11.896831512451172,
+      "learning_rate": 2e-05,
+      "loss": 0.8648,
+      "step": 100
+    },
+    {
+      "epoch": 0.18318068276436303,
+      "grad_norm": 10.90375804901123,
+      "learning_rate": 2.1800000000000005e-05,
+      "loss": 0.7764,
+      "step": 110
+    },
+    {
+      "epoch": 0.19983347210657784,
+      "grad_norm": 7.848816394805908,
+      "learning_rate": 2.3800000000000003e-05,
+      "loss": 0.6304,
+      "step": 120
+    },
+    {
+      "epoch": 0.21648626144879268,
+      "grad_norm": 5.973367214202881,
+      "learning_rate": 2.5800000000000004e-05,
+      "loss": 0.5106,
+      "step": 130
+    },
+    {
+      "epoch": 0.2331390507910075,
+      "grad_norm": 10.721240997314453,
+      "learning_rate": 2.78e-05,
+      "loss": 0.4462,
+      "step": 140
+    },
+    {
+      "epoch": 0.2497918401332223,
+      "grad_norm": 5.517733097076416,
+      "learning_rate": 2.9800000000000003e-05,
+      "loss": 0.5038,
+      "step": 150
+    },
+    {
+      "epoch": 0.2664446294754371,
+      "grad_norm": 8.605814933776855,
+      "learning_rate": 3.180000000000001e-05,
+      "loss": 0.3833,
+      "step": 160
+    },
+    {
+      "epoch": 0.28309741881765194,
+      "grad_norm": 1.7152849435806274,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.3112,
+      "step": 170
+    },
+    {
+      "epoch": 0.2997502081598668,
+      "grad_norm": 1.6370782852172852,
+      "learning_rate": 3.5600000000000005e-05,
+      "loss": 0.3238,
+      "step": 180
+    },
+    {
+      "epoch": 0.3164029975020816,
+      "grad_norm": 20.556215286254883,
+      "learning_rate": 3.76e-05,
+      "loss": 0.3753,
+      "step": 190
+    },
+    {
+      "epoch": 0.33305578684429643,
+      "grad_norm": 1.986198902130127,
+      "learning_rate": 3.96e-05,
+      "loss": 0.2539,
+      "step": 200
+    },
+    {
+      "epoch": 0.34970857618651124,
+      "grad_norm": 9.711119651794434,
+      "learning_rate": 3.9885714285714284e-05,
+      "loss": 0.195,
+      "step": 210
+    },
+    {
+      "epoch": 0.36636136552872606,
+      "grad_norm": 22.053802490234375,
+      "learning_rate": 3.9742857142857146e-05,
+      "loss": 0.2251,
+      "step": 220
+    },
+    {
+      "epoch": 0.38301415487094087,
+      "grad_norm": 4.119320869445801,
+      "learning_rate": 3.96e-05,
+      "loss": 0.2491,
+      "step": 230
+    },
+    {
+      "epoch": 0.3996669442131557,
+      "grad_norm": 34.85157012939453,
+      "learning_rate": 3.945714285714286e-05,
+      "loss": 0.2676,
+      "step": 240
+    },
+    {
+      "epoch": 0.4163197335553705,
+      "grad_norm": 3.7660233974456787,
+      "learning_rate": 3.9314285714285716e-05,
+      "loss": 0.2795,
+      "step": 250
+    },
+    {
+      "epoch": 0.43297252289758537,
+      "grad_norm": 2.8417813777923584,
+      "learning_rate": 3.917142857142858e-05,
+      "loss": 0.4349,
+      "step": 260
+    },
+    {
+      "epoch": 0.4496253122398002,
+      "grad_norm": 2.4724223613739014,
+      "learning_rate": 3.902857142857143e-05,
+      "loss": 0.2017,
+      "step": 270
+    },
+    {
+      "epoch": 0.466278101582015,
+      "grad_norm": 11.620677947998047,
+      "learning_rate": 3.8885714285714294e-05,
+      "loss": 0.1538,
+      "step": 280
+    },
+    {
+      "epoch": 0.4829308909242298,
+      "grad_norm": 8.653327941894531,
+      "learning_rate": 3.874285714285714e-05,
+      "loss": 0.3721,
+      "step": 290
+    },
+    {
+      "epoch": 0.4995836802664446,
+      "grad_norm": 0.48509305715560913,
+      "learning_rate": 3.86e-05,
+      "loss": 0.12,
+      "step": 300
+    },
+    {
+      "epoch": 0.5162364696086594,
+      "grad_norm": 16.0579833984375,
+      "learning_rate": 3.845714285714286e-05,
+      "loss": 0.1726,
+      "step": 310
+    },
+    {
+      "epoch": 0.5328892589508742,
+      "grad_norm": 11.486455917358398,
+      "learning_rate": 3.831428571428572e-05,
+      "loss": 0.1061,
+      "step": 320
+    },
+    {
+      "epoch": 0.5495420482930891,
+      "grad_norm": 9.900602340698242,
+      "learning_rate": 3.8171428571428574e-05,
+      "loss": 0.0963,
+      "step": 330
+    },
+    {
+      "epoch": 0.5661948376353039,
+      "grad_norm": 0.6496172547340393,
+      "learning_rate": 3.8028571428571435e-05,
+      "loss": 0.1222,
+      "step": 340
+    },
+    {
+      "epoch": 0.5828476269775187,
+      "grad_norm": 3.4763410091400146,
+      "learning_rate": 3.788571428571429e-05,
+      "loss": 0.0951,
+      "step": 350
+    },
+    {
+      "epoch": 0.5995004163197336,
+      "grad_norm": 1.4176121950149536,
+      "learning_rate": 3.7742857142857145e-05,
+      "loss": 0.1592,
+      "step": 360
+    },
+    {
+      "epoch": 0.6161532056619484,
+      "grad_norm": 0.24932004511356354,
+      "learning_rate": 3.76e-05,
+      "loss": 0.0837,
+      "step": 370
+    },
+    {
+      "epoch": 0.6328059950041632,
+      "grad_norm": 0.34060370922088623,
+      "learning_rate": 3.745714285714286e-05,
+      "loss": 0.1331,
+      "step": 380
+    },
+    {
+      "epoch": 0.649458784346378,
+      "grad_norm": 32.56100082397461,
+      "learning_rate": 3.7314285714285715e-05,
+      "loss": 0.0939,
+      "step": 390
+    },
+    {
+      "epoch": 0.6661115736885929,
+      "grad_norm": 27.72275733947754,
+      "learning_rate": 3.717142857142858e-05,
+      "loss": 0.2595,
+      "step": 400
+    },
+    {
+      "epoch": 0.6827643630308077,
+      "grad_norm": 2.5867972373962402,
+      "learning_rate": 3.702857142857143e-05,
+      "loss": 0.1213,
+      "step": 410
+    },
+    {
+      "epoch": 0.6994171523730225,
+      "grad_norm": 3.271141290664673,
+      "learning_rate": 3.6885714285714286e-05,
+      "loss": 0.0732,
+      "step": 420
+    },
+    {
+      "epoch": 0.7160699417152373,
+      "grad_norm": 0.40355125069618225,
+      "learning_rate": 3.674285714285715e-05,
+      "loss": 0.0865,
+      "step": 430
+    },
+    {
+      "epoch": 0.7327227310574521,
+      "grad_norm": 0.1593771129846573,
+      "learning_rate": 3.66e-05,
+      "loss": 0.0725,
+      "step": 440
+    },
+    {
+      "epoch": 0.7493755203996669,
+      "grad_norm": 6.7296857833862305,
+      "learning_rate": 3.645714285714286e-05,
+      "loss": 0.1532,
+      "step": 450
+    },
+    {
+      "epoch": 0.7660283097418817,
+      "grad_norm": 0.15103112161159515,
+      "learning_rate": 3.631428571428572e-05,
+      "loss": 0.0429,
+      "step": 460
+    },
+    {
+      "epoch": 0.7826810990840966,
+      "grad_norm": 10.738967895507812,
+      "learning_rate": 3.617142857142857e-05,
+      "loss": 0.0405,
+      "step": 470
+    },
+    {
+      "epoch": 0.7993338884263114,
+      "grad_norm": 0.13244017958641052,
+      "learning_rate": 3.602857142857143e-05,
+      "loss": 0.089,
+      "step": 480
+    },
+    {
+      "epoch": 0.8159866777685262,
+      "grad_norm": 0.13044649362564087,
+      "learning_rate": 3.588571428571429e-05,
+      "loss": 0.1373,
+      "step": 490
+    },
+    {
+      "epoch": 0.832639467110741,
+      "grad_norm": 0.2371596097946167,
+      "learning_rate": 3.574285714285714e-05,
+      "loss": 0.0845,
+      "step": 500
+    },
+    {
+      "epoch": 0.8492922564529559,
+      "grad_norm": 0.14471735060214996,
+      "learning_rate": 3.5600000000000005e-05,
+      "loss": 0.0872,
+      "step": 510
+    },
+    {
+      "epoch": 0.8659450457951707,
+      "grad_norm": 0.8515125513076782,
+      "learning_rate": 3.545714285714286e-05,
+      "loss": 0.0425,
+      "step": 520
+    },
+    {
+      "epoch": 0.8825978351373855,
+      "grad_norm": 1.3742948770523071,
+      "learning_rate": 3.531428571428572e-05,
+      "loss": 0.0114,
+      "step": 530
+    },
+    {
+      "epoch": 0.8992506244796004,
+      "grad_norm": 8.960271835327148,
+      "learning_rate": 3.5171428571428575e-05,
+      "loss": 0.0661,
+      "step": 540
+    },
+    {
+      "epoch": 0.9159034138218152,
+      "grad_norm": 0.16396482288837433,
+      "learning_rate": 3.502857142857143e-05,
+      "loss": 0.096,
+      "step": 550
+    },
+    {
+      "epoch": 0.93255620316403,
+      "grad_norm": 0.11441925913095474,
+      "learning_rate": 3.4885714285714285e-05,
+      "loss": 0.042,
+      "step": 560
+    },
+    {
+      "epoch": 0.9492089925062448,
+      "grad_norm": 0.9531962275505066,
+      "learning_rate": 3.4742857142857146e-05,
+      "loss": 0.1096,
+      "step": 570
+    },
+    {
+      "epoch": 0.9658617818484596,
+      "grad_norm": 20.664161682128906,
+      "learning_rate": 3.46e-05,
+      "loss": 0.1895,
+      "step": 580
+    },
+    {
+      "epoch": 0.9825145711906744,
+      "grad_norm": 0.10766961425542831,
+      "learning_rate": 3.445714285714286e-05,
+      "loss": 0.0951,
+      "step": 590
+    },
+    {
+      "epoch": 0.9991673605328892,
+      "grad_norm": 0.06997501850128174,
+      "learning_rate": 3.431428571428572e-05,
+      "loss": 0.0677,
+      "step": 600
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9935,
+      "eval_loss": 0.02969062514603138,
+      "eval_runtime": 43.3836,
+      "eval_samples_per_second": 46.1,
+      "eval_steps_per_second": 11.525,
+      "step": 601
+    },
+    {
+      "epoch": 1.0149875104079933,
+      "grad_norm": 0.9687513709068298,
+      "learning_rate": 3.417142857142858e-05,
+      "loss": 0.0525,
+      "step": 610
+    },
+    {
+      "epoch": 1.031640299750208,
+      "grad_norm": 6.680837154388428,
+      "learning_rate": 3.402857142857143e-05,
+      "loss": 0.0161,
+      "step": 620
+    },
+    {
+      "epoch": 1.048293089092423,
+      "grad_norm": 24.152660369873047,
+      "learning_rate": 3.388571428571429e-05,
+      "loss": 0.0387,
+      "step": 630
+    },
+    {
+      "epoch": 1.0649458784346377,
+      "grad_norm": 0.07758224755525589,
+      "learning_rate": 3.374285714285714e-05,
+      "loss": 0.0266,
+      "step": 640
+    },
+    {
+      "epoch": 1.0815986677768525,
+      "grad_norm": 0.0564408153295517,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.0563,
+      "step": 650
+    },
+    {
+      "epoch": 1.0982514571190674,
+      "grad_norm": 0.05482952296733856,
+      "learning_rate": 3.345714285714286e-05,
+      "loss": 0.0493,
+      "step": 660
+    },
+    {
+      "epoch": 1.1149042464612822,
+      "grad_norm": 0.05704626441001892,
+      "learning_rate": 3.331428571428572e-05,
+      "loss": 0.0924,
+      "step": 670
+    },
+    {
+      "epoch": 1.131557035803497,
+      "grad_norm": 0.05357396975159645,
+      "learning_rate": 3.3171428571428574e-05,
+      "loss": 0.0814,
+      "step": 680
+    },
+    {
+      "epoch": 1.1482098251457118,
+      "grad_norm": 0.47527748346328735,
+      "learning_rate": 3.302857142857143e-05,
+      "loss": 0.0544,
+      "step": 690
+    },
+    {
+      "epoch": 1.1648626144879266,
+      "grad_norm": 28.22191047668457,
+      "learning_rate": 3.288571428571429e-05,
+      "loss": 0.1022,
+      "step": 700
+    },
+    {
+      "epoch": 1.1815154038301416,
+      "grad_norm": 0.3312930166721344,
+      "learning_rate": 3.2742857142857145e-05,
+      "loss": 0.005,
+      "step": 710
+    },
+    {
+      "epoch": 1.1981681931723565,
+      "grad_norm": 35.735198974609375,
+      "learning_rate": 3.26e-05,
+      "loss": 0.0072,
+      "step": 720
+    },
+    {
+      "epoch": 1.2148209825145713,
+      "grad_norm": 0.04014894366264343,
+      "learning_rate": 3.245714285714286e-05,
+      "loss": 0.0938,
+      "step": 730
+    },
+    {
+      "epoch": 1.231473771856786,
+      "grad_norm": 16.871463775634766,
+      "learning_rate": 3.2314285714285716e-05,
+      "loss": 0.0241,
+      "step": 740
+    },
+    {
+      "epoch": 1.248126561199001,
+      "grad_norm": 25.51210594177246,
+      "learning_rate": 3.217142857142857e-05,
+      "loss": 0.0322,
+      "step": 750
+    },
+    {
+      "epoch": 1.2647793505412157,
+      "grad_norm": 0.03736506775021553,
+      "learning_rate": 3.202857142857143e-05,
+      "loss": 0.0521,
+      "step": 760
+    },
+    {
+      "epoch": 1.2814321398834305,
+      "grad_norm": 3.5099713802337646,
+      "learning_rate": 3.1885714285714286e-05,
+      "loss": 0.0582,
+      "step": 770
+    },
+    {
+      "epoch": 1.2980849292256453,
+      "grad_norm": 0.054958537220954895,
+      "learning_rate": 3.174285714285715e-05,
+      "loss": 0.0836,
+      "step": 780
+    },
+    {
+      "epoch": 1.3147377185678601,
+      "grad_norm": 0.7841320633888245,
+      "learning_rate": 3.16e-05,
+      "loss": 0.0066,
+      "step": 790
+    },
+    {
+      "epoch": 1.331390507910075,
+      "grad_norm": 0.03516022861003876,
+      "learning_rate": 3.1457142857142864e-05,
+      "loss": 0.0657,
+      "step": 800
+    },
+    {
+      "epoch": 1.3480432972522898,
+      "grad_norm": 0.03558405488729477,
+      "learning_rate": 3.131428571428571e-05,
+      "loss": 0.0347,
+      "step": 810
+    },
+    {
+      "epoch": 1.3646960865945046,
+      "grad_norm": 0.14143311977386475,
+      "learning_rate": 3.117142857142857e-05,
+      "loss": 0.0028,
+      "step": 820
+    },
+    {
+      "epoch": 1.3813488759367194,
+      "grad_norm": 2.7062411308288574,
+      "learning_rate": 3.102857142857143e-05,
+      "loss": 0.0898,
+      "step": 830
+    },
+    {
+      "epoch": 1.3980016652789342,
+      "grad_norm": 0.03832915052771568,
+      "learning_rate": 3.088571428571429e-05,
+      "loss": 0.004,
+      "step": 840
+    },
+    {
+      "epoch": 1.414654454621149,
+      "grad_norm": 19.017581939697266,
+      "learning_rate": 3.0742857142857144e-05,
+      "loss": 0.0495,
+      "step": 850
+    },
+    {
+      "epoch": 1.4313072439633638,
+      "grad_norm": 19.32662582397461,
+      "learning_rate": 3.0600000000000005e-05,
+      "loss": 0.0808,
+      "step": 860
+    },
+    {
+      "epoch": 1.4479600333055787,
+      "grad_norm": 0.21114251017570496,
+      "learning_rate": 3.045714285714286e-05,
+      "loss": 0.0275,
+      "step": 870
+    },
+    {
+      "epoch": 1.4646128226477935,
+      "grad_norm": 4.781786918640137,
+      "learning_rate": 3.0314285714285718e-05,
+      "loss": 0.0483,
+      "step": 880
+    },
+    {
+      "epoch": 1.4812656119900083,
+      "grad_norm": 25.905826568603516,
+      "learning_rate": 3.0171428571428576e-05,
+      "loss": 0.1505,
+      "step": 890
+    },
+    {
+      "epoch": 1.497918401332223,
+      "grad_norm": 8.17207145690918,
+      "learning_rate": 3.002857142857143e-05,
+      "loss": 0.0302,
+      "step": 900
+    },
+    {
+      "epoch": 1.5145711906744381,
+      "grad_norm": Infinity,
+      "learning_rate": 2.9900000000000005e-05,
+      "loss": 0.0906,
+      "step": 910
+    },
+    {
+      "epoch": 1.531223980016653,
+      "grad_norm": 3.578585624694824,
+      "learning_rate": 2.9757142857142856e-05,
+      "loss": 0.047,
+      "step": 920
+    },
+    {
+      "epoch": 1.5478767693588678,
+      "grad_norm": 0.03926655277609825,
+      "learning_rate": 2.9614285714285714e-05,
+      "loss": 0.0685,
+      "step": 930
+    },
+    {
+      "epoch": 1.5645295587010826,
+      "grad_norm": 30.002620697021484,
+      "learning_rate": 2.9471428571428572e-05,
+      "loss": 0.0474,
+      "step": 940
+    },
+    {
+      "epoch": 1.5811823480432974,
+      "grad_norm": 0.035517849028110504,
+      "learning_rate": 2.932857142857143e-05,
+      "loss": 0.0041,
+      "step": 950
+    },
+    {
+      "epoch": 1.5978351373855122,
+      "grad_norm": 4.368102550506592,
+      "learning_rate": 2.918571428571429e-05,
+      "loss": 0.0803,
+      "step": 960
+    },
+    {
+      "epoch": 1.614487926727727,
+      "grad_norm": 0.03676709160208702,
+      "learning_rate": 2.9042857142857146e-05,
+      "loss": 0.0037,
+      "step": 970
+    },
+    {
+      "epoch": 1.6311407160699418,
+      "grad_norm": 0.032644957304000854,
+      "learning_rate": 2.8900000000000005e-05,
+      "loss": 0.0379,
+      "step": 980
+    },
+    {
+      "epoch": 1.6477935054121566,
+      "grad_norm": 0.02497854270040989,
+      "learning_rate": 2.8757142857142863e-05,
+      "loss": 0.021,
+      "step": 990
+    },
+    {
+      "epoch": 1.6644462947543714,
+      "grad_norm": 0.024164369329810143,
+      "learning_rate": 2.8614285714285714e-05,
+      "loss": 0.012,
+      "step": 1000
+    },
+    {
+      "epoch": 1.6810990840965863,
+      "grad_norm": 0.05042396858334541,
+      "learning_rate": 2.8471428571428572e-05,
+      "loss": 0.0342,
+      "step": 1010
+    },
+    {
+      "epoch": 1.697751873438801,
+      "grad_norm": 0.15826770663261414,
+      "learning_rate": 2.832857142857143e-05,
+      "loss": 0.022,
+      "step": 1020
+    },
+    {
+      "epoch": 1.7144046627810159,
+      "grad_norm": 13.766159057617188,
+      "learning_rate": 2.8185714285714288e-05,
+      "loss": 0.0442,
+      "step": 1030
+    },
+    {
+      "epoch": 1.7310574521232307,
+      "grad_norm": 0.3997497856616974,
+      "learning_rate": 2.8042857142857146e-05,
+      "loss": 0.1228,
+      "step": 1040
+    },
+    {
+      "epoch": 1.7477102414654455,
+      "grad_norm": 0.04381551966071129,
+      "learning_rate": 2.7900000000000004e-05,
+      "loss": 0.0231,
+      "step": 1050
+    },
+    {
+      "epoch": 1.7643630308076603,
+      "grad_norm": 0.14710094034671783,
+      "learning_rate": 2.7757142857142862e-05,
+      "loss": 0.0995,
+      "step": 1060
+    },
+    {
+      "epoch": 1.7810158201498751,
+      "grad_norm": 4.448687553405762,
+      "learning_rate": 2.7614285714285717e-05,
+      "loss": 0.1546,
+      "step": 1070
+    },
+    {
+      "epoch": 1.79766860949209,
+      "grad_norm": 0.03282266855239868,
+      "learning_rate": 2.7471428571428575e-05,
+      "loss": 0.005,
+      "step": 1080
+    },
+    {
+      "epoch": 1.8143213988343048,
+      "grad_norm": 0.02770112454891205,
+      "learning_rate": 2.732857142857143e-05,
+      "loss": 0.009,
+      "step": 1090
+    },
+    {
+      "epoch": 1.8309741881765196,
+      "grad_norm": 0.02363790012896061,
+      "learning_rate": 2.7185714285714287e-05,
+      "loss": 0.0019,
+      "step": 1100
+    },
+    {
+      "epoch": 1.8476269775187344,
+      "grad_norm": 14.821802139282227,
+      "learning_rate": 2.7042857142857145e-05,
+      "loss": 0.028,
+      "step": 1110
+    },
+    {
+      "epoch": 1.8642797668609492,
+      "grad_norm": 0.024729197844862938,
+      "learning_rate": 2.6900000000000003e-05,
+      "loss": 0.0446,
+      "step": 1120
+    },
+    {
+      "epoch": 1.880932556203164,
+      "grad_norm": 0.039613548666238785,
+      "learning_rate": 2.6757142857142858e-05,
+      "loss": 0.0019,
+      "step": 1130
+    },
+    {
+      "epoch": 1.8975853455453788,
+      "grad_norm": 0.029998844489455223,
+      "learning_rate": 2.6614285714285716e-05,
+      "loss": 0.069,
+      "step": 1140
+    },
+    {
+      "epoch": 1.9142381348875936,
+      "grad_norm": 0.024138517677783966,
+      "learning_rate": 2.6471428571428574e-05,
+      "loss": 0.0708,
+      "step": 1150
+    },
+    {
+      "epoch": 1.9308909242298085,
+      "grad_norm": 0.02379235252737999,
+      "learning_rate": 2.6328571428571432e-05,
+      "loss": 0.0741,
+      "step": 1160
+    },
+    {
+      "epoch": 1.9475437135720233,
+      "grad_norm": 0.718246579170227,
+      "learning_rate": 2.618571428571429e-05,
+      "loss": 0.023,
+      "step": 1170
+    },
+    {
+      "epoch": 1.964196502914238,
+      "grad_norm": 0.021009549498558044,
+      "learning_rate": 2.6042857142857145e-05,
+      "loss": 0.0765,
+      "step": 1180
+    },
+    {
+      "epoch": 1.980849292256453,
+      "grad_norm": 7.245909690856934,
+      "learning_rate": 2.59e-05,
+      "loss": 0.086,
+      "step": 1190
+    },
+    {
+      "epoch": 1.9975020815986677,
+      "grad_norm": 0.024452045559883118,
+      "learning_rate": 2.5757142857142857e-05,
+      "loss": 0.0345,
+      "step": 1200
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9935,
+      "eval_loss": 0.03615127503871918,
+      "eval_runtime": 41.5933,
+      "eval_samples_per_second": 48.085,
+      "eval_steps_per_second": 12.021,
+      "step": 1202
+    },
+    {
+      "epoch": 2.0133222314737718,
+      "grad_norm": 10.592672348022461,
+      "learning_rate": 2.5614285714285715e-05,
+      "loss": 0.0139,
+      "step": 1210
+    },
+    {
+      "epoch": 2.0299750208159866,
+      "grad_norm": 0.06319635361433029,
+      "learning_rate": 2.5471428571428573e-05,
+      "loss": 0.0204,
+      "step": 1220
+    },
+    {
+      "epoch": 2.0466278101582014,
+      "grad_norm": 0.018913447856903076,
+      "learning_rate": 2.532857142857143e-05,
+      "loss": 0.0278,
+      "step": 1230
+    },
+    {
+      "epoch": 2.063280599500416,
+      "grad_norm": 0.019886815920472145,
+      "learning_rate": 2.518571428571429e-05,
+      "loss": 0.0195,
+      "step": 1240
+    },
+    {
+      "epoch": 2.079933388842631,
+      "grad_norm": 0.11551292985677719,
+      "learning_rate": 2.5042857142857148e-05,
+      "loss": 0.0349,
+      "step": 1250
+    },
+    {
+      "epoch": 2.096586178184846,
+      "grad_norm": 0.3485487103462219,
+      "learning_rate": 2.4900000000000006e-05,
+      "loss": 0.0084,
+      "step": 1260
+    },
+    {
+      "epoch": 2.1132389675270606,
+      "grad_norm": 0.040086254477500916,
+      "learning_rate": 2.4757142857142857e-05,
+      "loss": 0.0307,
+      "step": 1270
+    },
+    {
+      "epoch": 2.1298917568692755,
+      "grad_norm": 0.017919067293405533,
+      "learning_rate": 2.4614285714285715e-05,
+      "loss": 0.041,
+      "step": 1280
+    },
+    {
+      "epoch": 2.1465445462114903,
+      "grad_norm": 0.01724259741604328,
+      "learning_rate": 2.4471428571428573e-05,
+      "loss": 0.0052,
+      "step": 1290
+    },
+    {
+      "epoch": 2.163197335553705,
+      "grad_norm": 0.25569283962249756,
+      "learning_rate": 2.432857142857143e-05,
+      "loss": 0.0283,
+      "step": 1300
+    },
+    {
+      "epoch": 2.17985012489592,
+      "grad_norm": 0.016946446150541306,
+      "learning_rate": 2.418571428571429e-05,
+      "loss": 0.0124,
+      "step": 1310
+    },
+    {
+      "epoch": 2.1965029142381347,
+      "grad_norm": 0.023746447637677193,
+      "learning_rate": 2.4042857142857147e-05,
+      "loss": 0.0302,
+      "step": 1320
+    },
+    {
+      "epoch": 2.2131557035803495,
+      "grad_norm": 1.8454148769378662,
+      "learning_rate": 2.39e-05,
+      "loss": 0.0034,
+      "step": 1330
+    },
+    {
+      "epoch": 2.2298084929225643,
+      "grad_norm": 0.015741076320409775,
+      "learning_rate": 2.375714285714286e-05,
+      "loss": 0.0367,
+      "step": 1340
+    },
+    {
+      "epoch": 2.246461282264779,
+      "grad_norm": 0.015740349888801575,
+      "learning_rate": 2.3614285714285714e-05,
+      "loss": 0.0108,
+      "step": 1350
+    },
+    {
+      "epoch": 2.263114071606994,
+      "grad_norm": 20.47581672668457,
+      "learning_rate": 2.3471428571428572e-05,
+      "loss": 0.0652,
+      "step": 1360
+    },
+    {
+      "epoch": 2.279766860949209,
+      "grad_norm": 0.017650267109274864,
+      "learning_rate": 2.332857142857143e-05,
+      "loss": 0.0058,
+      "step": 1370
+    },
+    {
+      "epoch": 2.2964196502914236,
+      "grad_norm": 0.1071343645453453,
+      "learning_rate": 2.318571428571429e-05,
+      "loss": 0.0279,
+      "step": 1380
+    },
+    {
+      "epoch": 2.313072439633639,
+      "grad_norm": 0.16622492671012878,
+      "learning_rate": 2.3042857142857143e-05,
+      "loss": 0.0343,
+      "step": 1390
+    },
+    {
+      "epoch": 2.329725228975853,
+      "grad_norm": 0.07910218089818954,
+      "learning_rate": 2.29e-05,
+      "loss": 0.0023,
+      "step": 1400
+    },
+    {
+      "epoch": 2.3463780183180685,
+      "grad_norm": 0.016809897497296333,
+      "learning_rate": 2.275714285714286e-05,
+      "loss": 0.0081,
+      "step": 1410
+    },
+    {
+      "epoch": 2.3630308076602833,
+      "grad_norm": 0.014708627946674824,
+      "learning_rate": 2.2614285714285717e-05,
+      "loss": 0.0628,
+      "step": 1420
+    },
+    {
+      "epoch": 2.379683597002498,
+      "grad_norm": 0.014496504329144955,
+      "learning_rate": 2.2471428571428575e-05,
+      "loss": 0.0017,
+      "step": 1430
+    },
+    {
+      "epoch": 2.396336386344713,
+      "grad_norm": 0.013988097198307514,
+      "learning_rate": 2.232857142857143e-05,
+      "loss": 0.049,
+      "step": 1440
+    },
+    {
+      "epoch": 2.4129891756869277,
+      "grad_norm": 0.02206108532845974,
+      "learning_rate": 2.2185714285714284e-05,
+      "loss": 0.0009,
+      "step": 1450
+    },
+    {
+      "epoch": 2.4296419650291425,
+      "grad_norm": 0.01265011541545391,
+      "learning_rate": 2.2042857142857142e-05,
+      "loss": 0.0009,
+      "step": 1460
+    },
+    {
+      "epoch": 2.4462947543713573,
+      "grad_norm": 0.014575159177184105,
+      "learning_rate": 2.19e-05,
+      "loss": 0.0008,
+      "step": 1470
+    },
+    {
+      "epoch": 2.462947543713572,
+      "grad_norm": 0.012796576134860516,
+      "learning_rate": 2.175714285714286e-05,
+      "loss": 0.0425,
+      "step": 1480
+    },
+    {
+      "epoch": 2.479600333055787,
+      "grad_norm": 0.013447301462292671,
+      "learning_rate": 2.1614285714285716e-05,
+      "loss": 0.0011,
+      "step": 1490
+    },
+    {
+      "epoch": 2.496253122398002,
+      "grad_norm": 0.012052874080836773,
+      "learning_rate": 2.1471428571428574e-05,
+      "loss": 0.0153,
+      "step": 1500
+    },
+    {
+      "epoch": 2.5129059117402166,
+      "grad_norm": 0.011766649782657623,
+      "learning_rate": 2.1328571428571432e-05,
+      "loss": 0.0165,
+      "step": 1510
+    },
+    {
+      "epoch": 2.5295587010824314,
+      "grad_norm": 0.01666083373129368,
+      "learning_rate": 2.118571428571429e-05,
+      "loss": 0.0039,
+      "step": 1520
+    },
+    {
+      "epoch": 2.5462114904246462,
+      "grad_norm": 1.2848758697509766,
+      "learning_rate": 2.1042857142857142e-05,
+      "loss": 0.0383,
+      "step": 1530
+    },
+    {
+      "epoch": 2.562864279766861,
+      "grad_norm": 0.014356808736920357,
+      "learning_rate": 2.09e-05,
+      "loss": 0.0171,
+      "step": 1540
+    },
+    {
+      "epoch": 2.579517069109076,
+      "grad_norm": 0.08616916835308075,
+      "learning_rate": 2.0757142857142858e-05,
+      "loss": 0.0688,
+      "step": 1550
+    },
+    {
+      "epoch": 2.5961698584512907,
+      "grad_norm": 0.02040853165090084,
+      "learning_rate": 2.0614285714285716e-05,
+      "loss": 0.0011,
+      "step": 1560
+    },
+    {
+      "epoch": 2.6128226477935055,
+      "grad_norm": 15.67353343963623,
+      "learning_rate": 2.0471428571428574e-05,
+      "loss": 0.0273,
+      "step": 1570
+    },
+    {
+      "epoch": 2.6294754371357203,
+      "grad_norm": 0.014960126020014286,
+      "learning_rate": 2.0328571428571432e-05,
+      "loss": 0.0116,
+      "step": 1580
+    },
+    {
+      "epoch": 2.646128226477935,
+      "grad_norm": 0.0133629459887743,
+      "learning_rate": 2.018571428571429e-05,
+      "loss": 0.0154,
+      "step": 1590
+    },
+    {
+      "epoch": 2.66278101582015,
+      "grad_norm": 1.1109951734542847,
+      "learning_rate": 2.0042857142857145e-05,
+      "loss": 0.0298,
+      "step": 1600
+    },
+    {
+      "epoch": 2.6794338051623647,
+      "grad_norm": 0.07497254759073257,
+      "learning_rate": 1.9900000000000003e-05,
+      "loss": 0.0366,
+      "step": 1610
+    },
+    {
+      "epoch": 2.6960865945045795,
+      "grad_norm": 0.01391484122723341,
+      "learning_rate": 1.975714285714286e-05,
+      "loss": 0.0029,
+      "step": 1620
+    },
+    {
+      "epoch": 2.7127393838467944,
+      "grad_norm": 0.011711220256984234,
+      "learning_rate": 1.9614285714285715e-05,
+      "loss": 0.0183,
+      "step": 1630
+    },
+    {
+      "epoch": 2.729392173189009,
+      "grad_norm": 0.01194568071514368,
+      "learning_rate": 1.9471428571428573e-05,
+      "loss": 0.0232,
+      "step": 1640
+    },
+    {
+      "epoch": 2.746044962531224,
+      "grad_norm": 0.032725363969802856,
+      "learning_rate": 1.932857142857143e-05,
+      "loss": 0.0283,
+      "step": 1650
+    },
+    {
+      "epoch": 2.762697751873439,
+      "grad_norm": 0.021190594881772995,
+      "learning_rate": 1.9185714285714286e-05,
+      "loss": 0.0026,
+      "step": 1660
+    },
+    {
+      "epoch": 2.7793505412156536,
+      "grad_norm": 24.009925842285156,
+      "learning_rate": 1.9042857142857144e-05,
+      "loss": 0.0627,
+      "step": 1670
+    },
+    {
+      "epoch": 2.7960033305578684,
+      "grad_norm": 0.017221566289663315,
+      "learning_rate": 1.8900000000000002e-05,
+      "loss": 0.0132,
+      "step": 1680
+    },
+    {
+      "epoch": 2.8126561199000832,
+      "grad_norm": 0.010178760625422001,
+      "learning_rate": 1.8757142857142857e-05,
+      "loss": 0.0063,
+      "step": 1690
+    },
+    {
+      "epoch": 2.829308909242298,
+      "grad_norm": 0.01019757054746151,
+      "learning_rate": 1.8614285714285715e-05,
+      "loss": 0.0101,
+      "step": 1700
+    },
+    {
+      "epoch": 2.845961698584513,
+      "grad_norm": 0.010430095717310905,
+      "learning_rate": 1.8471428571428573e-05,
+      "loss": 0.0312,
+      "step": 1710
+    },
+    {
+      "epoch": 2.8626144879267277,
+      "grad_norm": 0.011290703900158405,
+      "learning_rate": 1.832857142857143e-05,
+      "loss": 0.0016,
+      "step": 1720
+    },
+    {
+      "epoch": 2.8792672772689425,
+      "grad_norm": 0.12969951331615448,
+      "learning_rate": 1.8185714285714285e-05,
+      "loss": 0.0354,
+      "step": 1730
+    },
+    {
+      "epoch": 2.8959200666111573,
+      "grad_norm": 0.011047742329537868,
+      "learning_rate": 1.8042857142857143e-05,
+      "loss": 0.0018,
+      "step": 1740
+    },
+    {
+      "epoch": 2.912572855953372,
+      "grad_norm": 0.010961545631289482,
+      "learning_rate": 1.79e-05,
+      "loss": 0.0023,
+      "step": 1750
+    },
+    {
+      "epoch": 2.929225645295587,
+      "grad_norm": 0.014039566740393639,
+      "learning_rate": 1.775714285714286e-05,
+      "loss": 0.0649,
+      "step": 1760
+    },
+    {
+      "epoch": 2.9458784346378017,
+      "grad_norm": 0.010026945732533932,
+      "learning_rate": 1.7614285714285714e-05,
+      "loss": 0.0039,
+      "step": 1770
+    },
+    {
+      "epoch": 2.9625312239800166,
+      "grad_norm": 0.009294411167502403,
+      "learning_rate": 1.7471428571428572e-05,
+      "loss": 0.0011,
+      "step": 1780
+    },
+    {
+      "epoch": 2.9791840133222314,
+      "grad_norm": 0.010262146592140198,
+      "learning_rate": 1.732857142857143e-05,
+      "loss": 0.0006,
+      "step": 1790
+    },
+    {
+      "epoch": 2.995836802664446,
+      "grad_norm": 19.167015075683594,
+      "learning_rate": 1.7185714285714288e-05,
+      "loss": 0.013,
+      "step": 1800
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.997,
+      "eval_loss": 0.015118513256311417,
+      "eval_runtime": 42.7084,
+      "eval_samples_per_second": 46.829,
+      "eval_steps_per_second": 11.707,
+      "step": 1803
+    },
+    {
+      "epoch": 3.0116569525395502,
+      "grad_norm": 0.0086745023727417,
+      "learning_rate": 1.7042857142857143e-05,
+      "loss": 0.0008,
+      "step": 1810
+    },
+    {
+      "epoch": 3.028309741881765,
+      "grad_norm": 0.17141631245613098,
+      "learning_rate": 1.69e-05,
+      "loss": 0.006,
+      "step": 1820
+    },
+    {
+      "epoch": 3.04496253122398,
+      "grad_norm": 0.01152071077376604,
+      "learning_rate": 1.675714285714286e-05,
+      "loss": 0.0006,
+      "step": 1830
+    },
+    {
+      "epoch": 3.0616153205661947,
+      "grad_norm": 0.00846625491976738,
+      "learning_rate": 1.6614285714285717e-05,
+      "loss": 0.0006,
+      "step": 1840
+    },
+    {
+      "epoch": 3.0782681099084095,
+      "grad_norm": 0.008341658860445023,
+      "learning_rate": 1.6471428571428575e-05,
+      "loss": 0.0005,
+      "step": 1850
+    },
+    {
+      "epoch": 3.0949208992506243,
+      "grad_norm": 0.008472333662211895,
+      "learning_rate": 1.632857142857143e-05,
+      "loss": 0.0005,
+      "step": 1860
+    },
+    {
+      "epoch": 3.111573688592839,
+      "grad_norm": 28.69985580444336,
+      "learning_rate": 1.6185714285714288e-05,
+      "loss": 0.009,
+      "step": 1870
+    },
+    {
+      "epoch": 3.128226477935054,
+      "grad_norm": 0.007999376393854618,
+      "learning_rate": 1.6042857142857146e-05,
+      "loss": 0.0005,
+      "step": 1880
+    },
+    {
+      "epoch": 3.1448792672772687,
+      "grad_norm": 0.008124138228595257,
+      "learning_rate": 1.5900000000000004e-05,
+      "loss": 0.0009,
+      "step": 1890
+    },
+    {
+      "epoch": 3.1615320566194836,
+      "grad_norm": 0.04555722326040268,
+      "learning_rate": 1.5757142857142858e-05,
+      "loss": 0.0137,
+      "step": 1900
+    },
+    {
+      "epoch": 3.178184845961699,
+      "grad_norm": 0.007663598284125328,
+      "learning_rate": 1.5614285714285716e-05,
+      "loss": 0.1333,
+      "step": 1910
+    },
+    {
+      "epoch": 3.194837635303913,
+      "grad_norm": 0.011646582745015621,
+      "learning_rate": 1.5471428571428574e-05,
+      "loss": 0.0506,
+      "step": 1920
+    },
+    {
+      "epoch": 3.2114904246461284,
+      "grad_norm": 0.00872163474559784,
+      "learning_rate": 1.532857142857143e-05,
+      "loss": 0.0037,
+      "step": 1930
+    },
+    {
+      "epoch": 3.2281432139883433,
+      "grad_norm": 0.008789711631834507,
+      "learning_rate": 1.5185714285714285e-05,
+      "loss": 0.0007,
+      "step": 1940
+    },
+    {
+      "epoch": 3.244796003330558,
+      "grad_norm": 0.008661571890115738,
+      "learning_rate": 1.5042857142857143e-05,
+      "loss": 0.0037,
+      "step": 1950
+    },
+    {
+      "epoch": 3.261448792672773,
+      "grad_norm": 28.83612060546875,
+      "learning_rate": 1.4900000000000001e-05,
+      "loss": 0.0253,
+      "step": 1960
+    },
+    {
+      "epoch": 3.2781015820149877,
+      "grad_norm": 0.0573546439409256,
+      "learning_rate": 1.475714285714286e-05,
+      "loss": 0.0007,
+      "step": 1970
+    },
+    {
+      "epoch": 3.2947543713572025,
+      "grad_norm": 0.009006750769913197,
+      "learning_rate": 1.4614285714285716e-05,
+      "loss": 0.0053,
+      "step": 1980
+    },
+    {
+      "epoch": 3.3114071606994173,
+      "grad_norm": 0.007715549319982529,
+      "learning_rate": 1.4471428571428572e-05,
+      "loss": 0.033,
+      "step": 1990
+    },
+    {
+      "epoch": 3.328059950041632,
+      "grad_norm": 0.008084608241915703,
+      "learning_rate": 1.432857142857143e-05,
+      "loss": 0.0144,
+      "step": 2000
+    },
+    {
+      "epoch": 3.344712739383847,
+      "grad_norm": 0.007967551238834858,
+      "learning_rate": 1.4185714285714286e-05,
+      "loss": 0.0137,
+      "step": 2010
+    },
+    {
+      "epoch": 3.3613655287260618,
+      "grad_norm": 0.008203152567148209,
+      "learning_rate": 1.4042857142857144e-05,
+      "loss": 0.0037,
+      "step": 2020
+    },
+    {
+      "epoch": 3.3780183180682766,
+      "grad_norm": 6.683016300201416,
+      "learning_rate": 1.39e-05,
+      "loss": 0.034,
+      "step": 2030
+    },
+    {
+      "epoch": 3.3946711074104914,
+      "grad_norm": 0.011201892048120499,
+      "learning_rate": 1.3757142857142857e-05,
+      "loss": 0.0005,
+      "step": 2040
+    },
+    {
+      "epoch": 3.411323896752706,
+      "grad_norm": 0.008145448751747608,
+      "learning_rate": 1.3614285714285715e-05,
+      "loss": 0.0005,
+      "step": 2050
+    },
+    {
+      "epoch": 3.427976686094921,
+      "grad_norm": 0.9828100204467773,
+      "learning_rate": 1.3471428571428573e-05,
+      "loss": 0.0007,
+      "step": 2060
+    },
+    {
+      "epoch": 3.444629475437136,
+      "grad_norm": 0.007419963832944632,
+      "learning_rate": 1.3328571428571431e-05,
+      "loss": 0.0124,
+      "step": 2070
+    },
+    {
+      "epoch": 3.4612822647793506,
+      "grad_norm": 0.030570199713110924,
+      "learning_rate": 1.3185714285714286e-05,
+      "loss": 0.0005,
+      "step": 2080
+    },
+    {
+      "epoch": 3.4779350541215655,
+      "grad_norm": 38.09405517578125,
+      "learning_rate": 1.3042857142857144e-05,
+      "loss": 0.0514,
+      "step": 2090
+    },
+    {
+      "epoch": 3.4945878434637803,
+      "grad_norm": 0.007145782001316547,
+      "learning_rate": 1.2900000000000002e-05,
+      "loss": 0.0005,
+      "step": 2100
+    },
+    {
+      "epoch": 3.511240632805995,
+      "grad_norm": 0.01062481477856636,
+      "learning_rate": 1.275714285714286e-05,
+      "loss": 0.0326,
+      "step": 2110
+    },
+    {
+      "epoch": 3.52789342214821,
+      "grad_norm": 0.00760676059871912,
+      "learning_rate": 1.2614285714285715e-05,
+      "loss": 0.0598,
+      "step": 2120
+    },
+    {
+      "epoch": 3.5445462114904247,
+      "grad_norm": 0.011588923633098602,
+      "learning_rate": 1.2471428571428573e-05,
+      "loss": 0.0317,
+      "step": 2130
+    },
+    {
+      "epoch": 3.5611990008326395,
+      "grad_norm": 0.007162110414355993,
+      "learning_rate": 1.232857142857143e-05,
+      "loss": 0.0007,
+      "step": 2140
+    },
+    {
+      "epoch": 3.5778517901748543,
+      "grad_norm": 0.008215065114200115,
+      "learning_rate": 1.2185714285714287e-05,
+      "loss": 0.0006,
+      "step": 2150
+    },
+    {
+      "epoch": 3.594504579517069,
+      "grad_norm": 0.34589236974716187,
+      "learning_rate": 1.2042857142857143e-05,
+      "loss": 0.0279,
+      "step": 2160
+    },
+    {
+      "epoch": 3.611157368859284,
+      "grad_norm": 0.02411896549165249,
+      "learning_rate": 1.1900000000000001e-05,
+      "loss": 0.0006,
+      "step": 2170
+    },
+    {
+      "epoch": 3.6278101582014988,
+      "grad_norm": 0.008000008761882782,
+      "learning_rate": 1.1757142857142858e-05,
+      "loss": 0.0232,
+      "step": 2180
+    },
+    {
+      "epoch": 3.6444629475437136,
+      "grad_norm": 0.01416712999343872,
+      "learning_rate": 1.1614285714285716e-05,
+      "loss": 0.0374,
+      "step": 2190
+    },
+    {
+      "epoch": 3.6611157368859284,
+      "grad_norm": 0.009390470571815968,
+      "learning_rate": 1.1471428571428574e-05,
+      "loss": 0.0132,
+      "step": 2200
+    },
+    {
+      "epoch": 3.677768526228143,
+      "grad_norm": 0.011863280087709427,
+      "learning_rate": 1.1328571428571428e-05,
+      "loss": 0.0008,
+      "step": 2210
+    },
+    {
+      "epoch": 3.694421315570358,
+      "grad_norm": 0.01807197742164135,
+      "learning_rate": 1.1185714285714286e-05,
+      "loss": 0.0006,
+      "step": 2220
+    },
+    {
+      "epoch": 3.711074104912573,
+      "grad_norm": 0.03740830719470978,
+      "learning_rate": 1.1042857142857144e-05,
+      "loss": 0.0006,
+      "step": 2230
+    },
+    {
+      "epoch": 3.7277268942547876,
+      "grad_norm": 0.007398474961519241,
+      "learning_rate": 1.0900000000000002e-05,
+      "loss": 0.0005,
+      "step": 2240
+    },
+    {
+      "epoch": 3.7443796835970025,
+      "grad_norm": 0.00720438826829195,
+      "learning_rate": 1.0757142857142857e-05,
+      "loss": 0.0005,
+      "step": 2250
+    },
+    {
+      "epoch": 3.7610324729392173,
+      "grad_norm": 0.007070284336805344,
+      "learning_rate": 1.0614285714285715e-05,
+      "loss": 0.0004,
+      "step": 2260
+    },
+    {
+      "epoch": 3.777685262281432,
+      "grad_norm": 0.007917001843452454,
+      "learning_rate": 1.0471428571428573e-05,
+      "loss": 0.0004,
+      "step": 2270
+    },
+    {
+      "epoch": 3.794338051623647,
+      "grad_norm": 0.006697102449834347,
+      "learning_rate": 1.032857142857143e-05,
+      "loss": 0.0138,
+      "step": 2280
+    },
+    {
+      "epoch": 3.8109908409658617,
+      "grad_norm": 0.008782276883721352,
+      "learning_rate": 1.0185714285714286e-05,
+      "loss": 0.0007,
+      "step": 2290
+    },
+    {
+      "epoch": 3.8276436303080765,
+      "grad_norm": 0.061034586280584335,
+      "learning_rate": 1.0042857142857144e-05,
+      "loss": 0.0005,
+      "step": 2300
+    },
+    {
+      "epoch": 3.8442964196502913,
+      "grad_norm": 0.006082494277507067,
+      "learning_rate": 9.9e-06,
+      "loss": 0.0004,
+      "step": 2310
+    },
+    {
+      "epoch": 3.860949208992506,
+      "grad_norm": 0.006540893577039242,
+      "learning_rate": 9.757142857142858e-06,
+      "loss": 0.0004,
+      "step": 2320
+    },
+    {
+      "epoch": 3.877601998334721,
+      "grad_norm": 0.0454992949962616,
+      "learning_rate": 9.614285714285714e-06,
+      "loss": 0.0004,
+      "step": 2330
+    },
+    {
+      "epoch": 3.894254787676936,
+      "grad_norm": 0.025658713653683662,
+      "learning_rate": 9.471428571428572e-06,
+      "loss": 0.0005,
+      "step": 2340
+    },
+    {
+      "epoch": 3.9109075770191506,
+      "grad_norm": 0.005761799868196249,
+      "learning_rate": 9.328571428571429e-06,
+      "loss": 0.0004,
+      "step": 2350
+    },
+    {
+      "epoch": 3.9275603663613654,
+      "grad_norm": 0.006150644738227129,
+      "learning_rate": 9.185714285714287e-06,
+      "loss": 0.0004,
+      "step": 2360
+    },
+    {
+      "epoch": 3.94421315570358,
+      "grad_norm": 0.0057544950395822525,
+      "learning_rate": 9.042857142857143e-06,
+      "loss": 0.0003,
+      "step": 2370
+    },
+    {
+      "epoch": 3.960865945045795,
+      "grad_norm": 0.006148363929241896,
+      "learning_rate": 8.900000000000001e-06,
+      "loss": 0.0248,
+      "step": 2380
+    },
+    {
+      "epoch": 3.97751873438801,
+      "grad_norm": 0.005814474541693926,
+      "learning_rate": 8.757142857142858e-06,
+      "loss": 0.0004,
+      "step": 2390
+    },
+    {
+      "epoch": 3.9941715237302247,
+      "grad_norm": 0.0060193813405931,
+      "learning_rate": 8.614285714285716e-06,
+      "loss": 0.0003,
+      "step": 2400
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.998,
+      "eval_loss": 0.013417072594165802,
+      "eval_runtime": 41.1291,
+      "eval_samples_per_second": 48.627,
+      "eval_steps_per_second": 12.157,
+      "step": 2404
+    },
+    {
+      "epoch": 4.009991673605329,
+      "grad_norm": 0.006395525299012661,
+      "learning_rate": 8.471428571428572e-06,
+      "loss": 0.0003,
+      "step": 2410
+    },
+    {
+      "epoch": 4.0266444629475435,
+      "grad_norm": 0.005516465287655592,
+      "learning_rate": 8.32857142857143e-06,
+      "loss": 0.0293,
+      "step": 2420
+    },
+    {
+      "epoch": 4.043297252289759,
+      "grad_norm": 0.5823869705200195,
+      "learning_rate": 8.185714285714286e-06,
+      "loss": 0.0004,
+      "step": 2430
+    },
+    {
+      "epoch": 4.059950041631973,
+      "grad_norm": 0.005808957852423191,
+      "learning_rate": 8.042857142857143e-06,
+      "loss": 0.0005,
+      "step": 2440
+    },
+    {
+      "epoch": 4.076602830974188,
+      "grad_norm": 0.00578009570017457,
+      "learning_rate": 7.9e-06,
+      "loss": 0.0003,
+      "step": 2450
+    },
+    {
+      "epoch": 4.093255620316403,
+      "grad_norm": 0.005703100468963385,
+      "learning_rate": 7.757142857142857e-06,
+      "loss": 0.0003,
+      "step": 2460
+    },
+    {
+      "epoch": 4.109908409658618,
+      "grad_norm": 0.005927779711782932,
+      "learning_rate": 7.614285714285715e-06,
+      "loss": 0.0003,
+      "step": 2470
+    },
+    {
+      "epoch": 4.126561199000832,
+      "grad_norm": 0.006832882761955261,
+      "learning_rate": 7.471428571428571e-06,
+      "loss": 0.0225,
+      "step": 2480
+    },
+    {
+      "epoch": 4.143213988343048,
+      "grad_norm": 0.005367404315620661,
+      "learning_rate": 7.328571428571429e-06,
+      "loss": 0.0008,
+      "step": 2490
+    },
+    {
+      "epoch": 4.159866777685262,
+      "grad_norm": 0.005708944518119097,
+      "learning_rate": 7.185714285714286e-06,
+      "loss": 0.038,
+      "step": 2500
+    },
+    {
+      "epoch": 4.176519567027477,
+      "grad_norm": 0.007482975255697966,
+      "learning_rate": 7.042857142857144e-06,
+      "loss": 0.0004,
+      "step": 2510
+    },
+    {
+      "epoch": 4.193172356369692,
+      "grad_norm": 0.007174965925514698,
+      "learning_rate": 6.9e-06,
+      "loss": 0.0006,
+      "step": 2520
+    },
+    {
+      "epoch": 4.209825145711907,
+      "grad_norm": 0.009364173747599125,
+      "learning_rate": 6.757142857142858e-06,
+      "loss": 0.0005,
+      "step": 2530
+    },
+    {
+      "epoch": 4.226477935054121,
+      "grad_norm": 0.007082544732838869,
+      "learning_rate": 6.614285714285715e-06,
+      "loss": 0.0258,
+      "step": 2540
+    },
+    {
+      "epoch": 4.2431307243963365,
+      "grad_norm": 0.006570044904947281,
+      "learning_rate": 6.4714285714285715e-06,
+      "loss": 0.0004,
+      "step": 2550
+    },
+    {
+      "epoch": 4.259783513738551,
+      "grad_norm": 0.006020836066454649,
+      "learning_rate": 6.3285714285714296e-06,
+      "loss": 0.0006,
+      "step": 2560
+    },
+    {
+      "epoch": 4.276436303080766,
+      "grad_norm": 0.005941543262451887,
+      "learning_rate": 6.185714285714286e-06,
+      "loss": 0.0009,
+      "step": 2570
+    },
+    {
+      "epoch": 4.2930890924229805,
+      "grad_norm": 0.0074948640540242195,
+      "learning_rate": 6.042857142857144e-06,
+      "loss": 0.0003,
+      "step": 2580
+    },
+    {
+      "epoch": 4.309741881765196,
+      "grad_norm": 0.0059341308660805225,
+      "learning_rate": 5.9e-06,
+      "loss": 0.0004,
+      "step": 2590
+    },
+    {
+      "epoch": 4.32639467110741,
+      "grad_norm": 0.00568350637331605,
+      "learning_rate": 5.7571428571428574e-06,
+      "loss": 0.0004,
+      "step": 2600
+    },
+    {
+      "epoch": 4.343047460449625,
+      "grad_norm": 0.005876988638192415,
+      "learning_rate": 5.614285714285715e-06,
+      "loss": 0.0311,
+      "step": 2610
+    },
+    {
+      "epoch": 4.35970024979184,
+      "grad_norm": 0.006036726292222738,
+      "learning_rate": 5.471428571428572e-06,
+      "loss": 0.0004,
+      "step": 2620
+    },
+    {
+      "epoch": 4.376353039134055,
+      "grad_norm": 0.013689212501049042,
+      "learning_rate": 5.328571428571428e-06,
+      "loss": 0.0017,
+      "step": 2630
+    },
+    {
+      "epoch": 4.393005828476269,
+      "grad_norm": 0.1334390938282013,
+      "learning_rate": 5.185714285714286e-06,
+      "loss": 0.0004,
+      "step": 2640
+    },
+    {
+      "epoch": 4.409658617818485,
+      "grad_norm": 0.023301932960748672,
+      "learning_rate": 5.042857142857144e-06,
+      "loss": 0.0004,
+      "step": 2650
+    },
+    {
+      "epoch": 4.426311407160699,
+      "grad_norm": 1.0275276899337769,
+      "learning_rate": 4.9000000000000005e-06,
+      "loss": 0.0004,
+      "step": 2660
+    },
+    {
+      "epoch": 4.442964196502914,
+      "grad_norm": 0.005459700245410204,
+      "learning_rate": 4.757142857142858e-06,
+      "loss": 0.0003,
+      "step": 2670
+    },
+    {
+      "epoch": 4.459616985845129,
+      "grad_norm": 0.019774090498685837,
+      "learning_rate": 4.614285714285715e-06,
+      "loss": 0.0327,
+      "step": 2680
+    },
+    {
+      "epoch": 4.476269775187344,
+      "grad_norm": 0.009837147779762745,
+      "learning_rate": 4.471428571428571e-06,
+      "loss": 0.0125,
+      "step": 2690
+    },
+    {
+      "epoch": 4.492922564529558,
+      "grad_norm": 0.005503670312464237,
+      "learning_rate": 4.328571428571429e-06,
+      "loss": 0.0003,
+      "step": 2700
+    },
+    {
+      "epoch": 4.5095753538717736,
+      "grad_norm": 0.0053068650886416435,
+      "learning_rate": 4.185714285714286e-06,
+      "loss": 0.0004,
+      "step": 2710
+    },
+    {
+      "epoch": 4.526228143213988,
+      "grad_norm": 0.005588301923125982,
+      "learning_rate": 4.042857142857144e-06,
+      "loss": 0.0222,
+      "step": 2720
+    },
+    {
+      "epoch": 4.542880932556203,
+      "grad_norm": 0.0052786581218242645,
+      "learning_rate": 3.900000000000001e-06,
+      "loss": 0.0003,
+      "step": 2730
+    },
+    {
+      "epoch": 4.559533721898418,
+      "grad_norm": 0.005984195042401552,
+      "learning_rate": 3.7571428571428575e-06,
+      "loss": 0.0003,
+      "step": 2740
+    },
+    {
+      "epoch": 4.576186511240633,
+      "grad_norm": 0.005208215676248074,
+      "learning_rate": 3.6142857142857147e-06,
+      "loss": 0.0003,
+      "step": 2750
+    },
+    {
+      "epoch": 4.592839300582847,
+      "grad_norm": 0.16855907440185547,
+      "learning_rate": 3.471428571428572e-06,
+      "loss": 0.0164,
+      "step": 2760
+    },
+    {
+      "epoch": 4.609492089925062,
+      "grad_norm": 0.006120254285633564,
+      "learning_rate": 3.3285714285714286e-06,
+      "loss": 0.0003,
+      "step": 2770
+    },
+    {
+      "epoch": 4.626144879267278,
+      "grad_norm": 0.013766973279416561,
+      "learning_rate": 3.185714285714286e-06,
+      "loss": 0.0003,
+      "step": 2780
+    },
+    {
+      "epoch": 4.642797668609492,
+      "grad_norm": 0.10941141843795776,
+      "learning_rate": 3.042857142857143e-06,
+      "loss": 0.0003,
+      "step": 2790
+    },
+    {
+      "epoch": 4.659450457951706,
+      "grad_norm": 0.005121996160596609,
+      "learning_rate": 2.9e-06,
+      "loss": 0.0003,
+      "step": 2800
+    },
+    {
+      "epoch": 4.676103247293922,
+      "grad_norm": 0.005419144406914711,
+      "learning_rate": 2.757142857142857e-06,
+      "loss": 0.0087,
+      "step": 2810
+    },
+    {
+      "epoch": 4.692756036636137,
+      "grad_norm": 0.005216961260885,
+      "learning_rate": 2.614285714285715e-06,
+      "loss": 0.0038,
+      "step": 2820
+    },
+    {
+      "epoch": 4.709408825978351,
+      "grad_norm": 0.005281897261738777,
+      "learning_rate": 2.4714285714285717e-06,
+      "loss": 0.0003,
+      "step": 2830
+    },
+    {
+      "epoch": 4.726061615320567,
+      "grad_norm": 0.005348918028175831,
+      "learning_rate": 2.3285714285714285e-06,
+      "loss": 0.0017,
+      "step": 2840
+    },
+    {
+      "epoch": 4.742714404662781,
+      "grad_norm": 0.005255383439362049,
+      "learning_rate": 2.185714285714286e-06,
+      "loss": 0.0003,
+      "step": 2850
+    },
+    {
+      "epoch": 4.759367194004996,
+      "grad_norm": 0.004928835202008486,
+      "learning_rate": 2.0428571428571433e-06,
+      "loss": 0.0003,
+      "step": 2860
+    },
+    {
+      "epoch": 4.776019983347211,
+      "grad_norm": 11.944090843200684,
+      "learning_rate": 1.9000000000000002e-06,
+      "loss": 0.0137,
+      "step": 2870
+    },
+    {
+      "epoch": 4.792672772689426,
+      "grad_norm": 0.0049188993871212006,
+      "learning_rate": 1.7571428571428572e-06,
+      "loss": 0.0004,
+      "step": 2880
+    },
+    {
+      "epoch": 4.80932556203164,
+      "grad_norm": 0.005385698284953833,
+      "learning_rate": 1.6142857142857144e-06,
+      "loss": 0.0003,
+      "step": 2890
+    },
+    {
+      "epoch": 4.8259783513738554,
+      "grad_norm": 0.0051418510265648365,
+      "learning_rate": 1.4714285714285713e-06,
+      "loss": 0.0004,
+      "step": 2900
+    },
+    {
+      "epoch": 4.84263114071607,
+      "grad_norm": 0.005018496885895729,
+      "learning_rate": 1.3285714285714287e-06,
+      "loss": 0.0004,
+      "step": 2910
+    },
+    {
+      "epoch": 4.859283930058285,
+      "grad_norm": 0.005061938893049955,
+      "learning_rate": 1.185714285714286e-06,
+      "loss": 0.0003,
+      "step": 2920
+    },
+    {
+      "epoch": 4.875936719400499,
+      "grad_norm": 0.0050169117748737335,
+      "learning_rate": 1.0428571428571429e-06,
+      "loss": 0.0003,
+      "step": 2930
+    },
+    {
+      "epoch": 4.892589508742715,
+      "grad_norm": 0.009882300160825253,
+      "learning_rate": 9.000000000000001e-07,
+      "loss": 0.0003,
+      "step": 2940
+    },
+    {
+      "epoch": 4.909242298084929,
+      "grad_norm": 0.005003802943974733,
+      "learning_rate": 7.571428571428572e-07,
+      "loss": 0.0003,
+      "step": 2950
+    },
+    {
+      "epoch": 4.925895087427144,
+      "grad_norm": 0.005712533835321665,
+      "learning_rate": 6.142857142857143e-07,
+      "loss": 0.0091,
+      "step": 2960
+    },
+    {
+      "epoch": 4.942547876769359,
+      "grad_norm": 0.005044932942837477,
+      "learning_rate": 4.714285714285715e-07,
+      "loss": 0.0004,
+      "step": 2970
+    },
+    {
+      "epoch": 4.959200666111574,
+      "grad_norm": 0.004893309436738491,
+      "learning_rate": 3.2857142857142857e-07,
+      "loss": 0.0003,
+      "step": 2980
+    },
+    {
+      "epoch": 4.975853455453788,
+      "grad_norm": 0.004968111868947744,
+      "learning_rate": 1.8571428571428572e-07,
+      "loss": 0.0003,
+      "step": 2990
+    },
+    {
+      "epoch": 4.992506244796004,
+      "grad_norm": 0.004901645239442587,
+      "learning_rate": 4.2857142857142865e-08,
+      "loss": 0.0003,
+      "step": 3000
+    },
+    {
+      "epoch": 4.992506244796004,
+      "eval_accuracy": 0.998,
+      "eval_loss": 0.012748559936881065,
+      "eval_runtime": 52.3226,
+      "eval_samples_per_second": 38.224,
+      "eval_steps_per_second": 9.556,
+      "step": 3000
+    },
+    {
+      "epoch": 4.992506244796004,
+      "step": 3000,
+      "total_flos": 1.2141162127800926e+19,
+      "train_loss": 0.10829704528961641,
+      "train_runtime": 6533.0547,
+      "train_samples_per_second": 22.052,
+      "train_steps_per_second": 0.459
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.2141162127800926e+19,
+  "train_batch_size": 24,
+  "trial_name": null,
+  "trial_params": null
+}