End of training

README.md

@@ -1,6 +1,7 @@
 ---
 base_model: anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test
 tags:
+- audio-classification
 - generated_from_trainer
 metrics:
 - accuracy
@@ -14,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-5Class-Validation-Mobil
 
-This model is a fine-tuned version of [anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test](https://huggingface.co/anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test) on an unknown dataset.
+This model is a fine-tuned version of [anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test](https://huggingface.co/anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test) on the anderloh/ValidateRes dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2506
+- Loss: 1.2514
 - Accuracy: 0.5836
 
 ## Model description

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 276.92,
+    "eval_accuracy": 0.5836298932384342,
+    "eval_loss": 1.2513927221298218,
+    "eval_runtime": 4.3951,
+    "eval_samples_per_second": 63.934,
+    "eval_steps_per_second": 0.683,
+    "train_loss": 0.8810926691691081,
+    "train_runtime": 3759.0782,
+    "train_samples_per_second": 123.541,
+    "train_steps_per_second": 0.239
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 276.92,
+    "eval_accuracy": 0.5836298932384342,
+    "eval_loss": 1.2513927221298218,
+    "eval_runtime": 4.3951,
+    "eval_samples_per_second": 63.934,
+    "eval_steps_per_second": 0.683
+}

runs/Jul13_20-32-33_ml6.hpc.uio.no/events.out.tfevents.1720899362.ml6.hpc.uio.no.4128034.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3778173a98ea9f7d3b60f4423177e2c14ca910de0ab74eb538c18836879e8e40
+size 411

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 276.92,
+    "train_loss": 0.8810926691691081,
+    "train_runtime": 3759.0782,
+    "train_samples_per_second": 123.541,
+    "train_steps_per_second": 0.239
+}

trainer_state.json

@@ -0,0 +1,2586 @@
+{
+  "best_metric": 0.5836298932384342,
+  "best_model_checkpoint": "wav2vec2-5Class-Validation-Mobil/checkpoint-773",
+  "epoch": 276.9230769230769,
+  "eval_steps": 500,
+  "global_step": 900,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.92,
+      "eval_accuracy": 0.3202846975088968,
+      "eval_loss": 1.602386713027954,
+      "eval_runtime": 4.3468,
+      "eval_samples_per_second": 64.645,
+      "eval_steps_per_second": 0.69,
+      "step": 3
+    },
+    {
+      "epoch": 1.85,
+      "eval_accuracy": 0.3167259786476868,
+      "eval_loss": 1.6022303104400635,
+      "eval_runtime": 3.573,
+      "eval_samples_per_second": 78.645,
+      "eval_steps_per_second": 0.84,
+      "step": 6
+    },
+    {
+      "epoch": 2.77,
+      "eval_accuracy": 0.3167259786476868,
+      "eval_loss": 1.601974368095398,
+      "eval_runtime": 4.6151,
+      "eval_samples_per_second": 60.887,
+      "eval_steps_per_second": 0.65,
+      "step": 9
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.3167259786476868,
+      "eval_loss": 1.6014597415924072,
+      "eval_runtime": 5.3659,
+      "eval_samples_per_second": 52.368,
+      "eval_steps_per_second": 0.559,
+      "step": 13
+    },
+    {
+      "epoch": 4.92,
+      "eval_accuracy": 0.3167259786476868,
+      "eval_loss": 1.6009386777877808,
+      "eval_runtime": 3.4504,
+      "eval_samples_per_second": 81.439,
+      "eval_steps_per_second": 0.869,
+      "step": 16
+    },
+    {
+      "epoch": 5.85,
+      "eval_accuracy": 0.31316725978647686,
+      "eval_loss": 1.6003268957138062,
+      "eval_runtime": 4.2937,
+      "eval_samples_per_second": 65.445,
+      "eval_steps_per_second": 0.699,
+      "step": 19
+    },
+    {
+      "epoch": 6.77,
+      "eval_accuracy": 0.30604982206405695,
+      "eval_loss": 1.5995941162109375,
+      "eval_runtime": 3.7057,
+      "eval_samples_per_second": 75.828,
+      "eval_steps_per_second": 0.81,
+      "step": 22
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.298932384341637,
+      "eval_loss": 1.5984183549880981,
+      "eval_runtime": 4.6458,
+      "eval_samples_per_second": 60.484,
+      "eval_steps_per_second": 0.646,
+      "step": 26
+    },
+    {
+      "epoch": 8.92,
+      "eval_accuracy": 0.2918149466192171,
+      "eval_loss": 1.5974235534667969,
+      "eval_runtime": 5.0303,
+      "eval_samples_per_second": 55.861,
+      "eval_steps_per_second": 0.596,
+      "step": 29
+    },
+    {
+      "epoch": 9.85,
+      "eval_accuracy": 0.27402135231316727,
+      "eval_loss": 1.596360445022583,
+      "eval_runtime": 3.3268,
+      "eval_samples_per_second": 84.465,
+      "eval_steps_per_second": 0.902,
+      "step": 32
+    },
+    {
+      "epoch": 10.77,
+      "eval_accuracy": 0.2597864768683274,
+      "eval_loss": 1.5951836109161377,
+      "eval_runtime": 3.1882,
+      "eval_samples_per_second": 88.138,
+      "eval_steps_per_second": 0.941,
+      "step": 35
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.26334519572953735,
+      "eval_loss": 1.593432903289795,
+      "eval_runtime": 4.2078,
+      "eval_samples_per_second": 66.78,
+      "eval_steps_per_second": 0.713,
+      "step": 39
+    },
+    {
+      "epoch": 12.92,
+      "eval_accuracy": 0.27402135231316727,
+      "eval_loss": 1.5920255184173584,
+      "eval_runtime": 4.9074,
+      "eval_samples_per_second": 57.261,
+      "eval_steps_per_second": 0.611,
+      "step": 42
+    },
+    {
+      "epoch": 13.85,
+      "eval_accuracy": 0.298932384341637,
+      "eval_loss": 1.5904992818832397,
+      "eval_runtime": 5.4737,
+      "eval_samples_per_second": 51.336,
+      "eval_steps_per_second": 0.548,
+      "step": 45
+    },
+    {
+      "epoch": 14.77,
+      "eval_accuracy": 0.298932384341637,
+      "eval_loss": 1.5889027118682861,
+      "eval_runtime": 5.4844,
+      "eval_samples_per_second": 51.236,
+      "eval_steps_per_second": 0.547,
+      "step": 48
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.2846975088967972,
+      "eval_loss": 1.5867795944213867,
+      "eval_runtime": 4.8027,
+      "eval_samples_per_second": 58.508,
+      "eval_steps_per_second": 0.625,
+      "step": 52
+    },
+    {
+      "epoch": 16.92,
+      "eval_accuracy": 0.2846975088967972,
+      "eval_loss": 1.5850844383239746,
+      "eval_runtime": 4.5938,
+      "eval_samples_per_second": 61.169,
+      "eval_steps_per_second": 0.653,
+      "step": 55
+    },
+    {
+      "epoch": 17.85,
+      "eval_accuracy": 0.2846975088967972,
+      "eval_loss": 1.5833449363708496,
+      "eval_runtime": 3.4722,
+      "eval_samples_per_second": 80.929,
+      "eval_steps_per_second": 0.864,
+      "step": 58
+    },
+    {
+      "epoch": 18.77,
+      "eval_accuracy": 0.26334519572953735,
+      "eval_loss": 1.58156418800354,
+      "eval_runtime": 3.9515,
+      "eval_samples_per_second": 71.112,
+      "eval_steps_per_second": 0.759,
+      "step": 61
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.24555160142348753,
+      "eval_loss": 1.579047441482544,
+      "eval_runtime": 4.2125,
+      "eval_samples_per_second": 66.707,
+      "eval_steps_per_second": 0.712,
+      "step": 65
+    },
+    {
+      "epoch": 20.92,
+      "eval_accuracy": 0.24199288256227758,
+      "eval_loss": 1.576985478401184,
+      "eval_runtime": 4.6275,
+      "eval_samples_per_second": 60.724,
+      "eval_steps_per_second": 0.648,
+      "step": 68
+    },
+    {
+      "epoch": 21.85,
+      "eval_accuracy": 0.23487544483985764,
+      "eval_loss": 1.574812650680542,
+      "eval_runtime": 4.9061,
+      "eval_samples_per_second": 57.275,
+      "eval_steps_per_second": 0.611,
+      "step": 71
+    },
+    {
+      "epoch": 22.77,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5727591514587402,
+      "eval_runtime": 5.6003,
+      "eval_samples_per_second": 50.176,
+      "eval_steps_per_second": 0.536,
+      "step": 74
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.2277580071174377,
+      "eval_loss": 1.5699430704116821,
+      "eval_runtime": 4.5057,
+      "eval_samples_per_second": 62.365,
+      "eval_steps_per_second": 0.666,
+      "step": 78
+    },
+    {
+      "epoch": 24.92,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.567823052406311,
+      "eval_runtime": 4.5731,
+      "eval_samples_per_second": 61.446,
+      "eval_steps_per_second": 0.656,
+      "step": 81
+    },
+    {
+      "epoch": 25.85,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5657496452331543,
+      "eval_runtime": 4.3556,
+      "eval_samples_per_second": 64.515,
+      "eval_steps_per_second": 0.689,
+      "step": 84
+    },
+    {
+      "epoch": 26.77,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5637929439544678,
+      "eval_runtime": 5.9441,
+      "eval_samples_per_second": 47.274,
+      "eval_steps_per_second": 0.505,
+      "step": 87
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5613017082214355,
+      "eval_runtime": 4.5762,
+      "eval_samples_per_second": 61.404,
+      "eval_steps_per_second": 0.656,
+      "step": 91
+    },
+    {
+      "epoch": 28.92,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5597190856933594,
+      "eval_runtime": 4.1813,
+      "eval_samples_per_second": 67.204,
+      "eval_steps_per_second": 0.717,
+      "step": 94
+    },
+    {
+      "epoch": 29.85,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5587605237960815,
+      "eval_runtime": 4.6749,
+      "eval_samples_per_second": 60.108,
+      "eval_steps_per_second": 0.642,
+      "step": 97
+    },
+    {
+      "epoch": 30.77,
+      "grad_norm": 66708.1953125,
+      "learning_rate": 2.962962962962963e-05,
+      "loss": 1.561,
+      "step": 100
+    },
+    {
+      "epoch": 30.77,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5586402416229248,
+      "eval_runtime": 5.2059,
+      "eval_samples_per_second": 53.977,
+      "eval_steps_per_second": 0.576,
+      "step": 100
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5596789121627808,
+      "eval_runtime": 4.428,
+      "eval_samples_per_second": 63.46,
+      "eval_steps_per_second": 0.678,
+      "step": 104
+    },
+    {
+      "epoch": 32.92,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5619100332260132,
+      "eval_runtime": 3.3009,
+      "eval_samples_per_second": 85.128,
+      "eval_steps_per_second": 0.909,
+      "step": 107
+    },
+    {
+      "epoch": 33.85,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5660569667816162,
+      "eval_runtime": 3.371,
+      "eval_samples_per_second": 83.357,
+      "eval_steps_per_second": 0.89,
+      "step": 110
+    },
+    {
+      "epoch": 34.77,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5720349550247192,
+      "eval_runtime": 3.9013,
+      "eval_samples_per_second": 72.028,
+      "eval_steps_per_second": 0.769,
+      "step": 113
+    },
+    {
+      "epoch": 36.0,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5833308696746826,
+      "eval_runtime": 4.7161,
+      "eval_samples_per_second": 59.583,
+      "eval_steps_per_second": 0.636,
+      "step": 117
+    },
+    {
+      "epoch": 36.92,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.5957212448120117,
+      "eval_runtime": 4.1977,
+      "eval_samples_per_second": 66.942,
+      "eval_steps_per_second": 0.715,
+      "step": 120
+    },
+    {
+      "epoch": 37.85,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.6119521856307983,
+      "eval_runtime": 3.034,
+      "eval_samples_per_second": 92.618,
+      "eval_steps_per_second": 0.989,
+      "step": 123
+    },
+    {
+      "epoch": 38.77,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.631814956665039,
+      "eval_runtime": 3.0252,
+      "eval_samples_per_second": 92.887,
+      "eval_steps_per_second": 0.992,
+      "step": 126
+    },
+    {
+      "epoch": 40.0,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.663757085800171,
+      "eval_runtime": 3.243,
+      "eval_samples_per_second": 86.648,
+      "eval_steps_per_second": 0.925,
+      "step": 130
+    },
+    {
+      "epoch": 40.92,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.6904593706130981,
+      "eval_runtime": 3.1943,
+      "eval_samples_per_second": 87.97,
+      "eval_steps_per_second": 0.939,
+      "step": 133
+    },
+    {
+      "epoch": 41.85,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.7196571826934814,
+      "eval_runtime": 3.4764,
+      "eval_samples_per_second": 80.832,
+      "eval_steps_per_second": 0.863,
+      "step": 136
+    },
+    {
+      "epoch": 42.77,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.750288724899292,
+      "eval_runtime": 3.415,
+      "eval_samples_per_second": 82.283,
+      "eval_steps_per_second": 0.878,
+      "step": 139
+    },
+    {
+      "epoch": 44.0,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.7802847623825073,
+      "eval_runtime": 3.0779,
+      "eval_samples_per_second": 91.295,
+      "eval_steps_per_second": 0.975,
+      "step": 143
+    },
+    {
+      "epoch": 44.92,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.7917312383651733,
+      "eval_runtime": 3.6229,
+      "eval_samples_per_second": 77.562,
+      "eval_steps_per_second": 0.828,
+      "step": 146
+    },
+    {
+      "epoch": 45.85,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.7919948101043701,
+      "eval_runtime": 3.2733,
+      "eval_samples_per_second": 85.845,
+      "eval_steps_per_second": 0.916,
+      "step": 149
+    },
+    {
+      "epoch": 46.77,
+      "eval_accuracy": 0.2313167259786477,
+      "eval_loss": 1.7869282960891724,
+      "eval_runtime": 3.1081,
+      "eval_samples_per_second": 90.408,
+      "eval_steps_per_second": 0.965,
+      "step": 152
+    },
+    {
+      "epoch": 48.0,
+      "eval_accuracy": 0.2597864768683274,
+      "eval_loss": 1.7699986696243286,
+      "eval_runtime": 3.2526,
+      "eval_samples_per_second": 86.392,
+      "eval_steps_per_second": 0.922,
+      "step": 156
+    },
+    {
+      "epoch": 48.92,
+      "eval_accuracy": 0.27402135231316727,
+      "eval_loss": 1.7525370121002197,
+      "eval_runtime": 2.789,
+      "eval_samples_per_second": 100.754,
+      "eval_steps_per_second": 1.076,
+      "step": 159
+    },
+    {
+      "epoch": 49.85,
+      "eval_accuracy": 0.2775800711743772,
+      "eval_loss": 1.7406829595565796,
+      "eval_runtime": 3.5203,
+      "eval_samples_per_second": 79.822,
+      "eval_steps_per_second": 0.852,
+      "step": 162
+    },
+    {
+      "epoch": 50.77,
+      "eval_accuracy": 0.2918149466192171,
+      "eval_loss": 1.7306878566741943,
+      "eval_runtime": 3.4092,
+      "eval_samples_per_second": 82.424,
+      "eval_steps_per_second": 0.88,
+      "step": 165
+    },
+    {
+      "epoch": 52.0,
+      "eval_accuracy": 0.3096085409252669,
+      "eval_loss": 1.7241473197937012,
+      "eval_runtime": 3.4771,
+      "eval_samples_per_second": 80.815,
+      "eval_steps_per_second": 0.863,
+      "step": 169
+    },
+    {
+      "epoch": 52.92,
+      "eval_accuracy": 0.3167259786476868,
+      "eval_loss": 1.7242671251296997,
+      "eval_runtime": 3.338,
+      "eval_samples_per_second": 84.182,
+      "eval_steps_per_second": 0.899,
+      "step": 172
+    },
+    {
+      "epoch": 53.85,
+      "eval_accuracy": 0.3167259786476868,
+      "eval_loss": 1.7253814935684204,
+      "eval_runtime": 3.037,
+      "eval_samples_per_second": 92.524,
+      "eval_steps_per_second": 0.988,
+      "step": 175
+    },
+    {
+      "epoch": 54.77,
+      "eval_accuracy": 0.3238434163701068,
+      "eval_loss": 1.7232733964920044,
+      "eval_runtime": 3.3453,
+      "eval_samples_per_second": 84.0,
+      "eval_steps_per_second": 0.897,
+      "step": 178
+    },
+    {
+      "epoch": 56.0,
+      "eval_accuracy": 0.3238434163701068,
+      "eval_loss": 1.7224737405776978,
+      "eval_runtime": 4.1856,
+      "eval_samples_per_second": 67.135,
+      "eval_steps_per_second": 0.717,
+      "step": 182
+    },
+    {
+      "epoch": 56.92,
+      "eval_accuracy": 0.3274021352313167,
+      "eval_loss": 1.7187089920043945,
+      "eval_runtime": 4.0825,
+      "eval_samples_per_second": 68.831,
+      "eval_steps_per_second": 0.735,
+      "step": 185
+    },
+    {
+      "epoch": 57.85,
+      "eval_accuracy": 0.3274021352313167,
+      "eval_loss": 1.7172435522079468,
+      "eval_runtime": 4.3988,
+      "eval_samples_per_second": 63.881,
+      "eval_steps_per_second": 0.682,
+      "step": 188
+    },
+    {
+      "epoch": 58.77,
+      "eval_accuracy": 0.33451957295373663,
+      "eval_loss": 1.7145518064498901,
+      "eval_runtime": 3.5886,
+      "eval_samples_per_second": 78.303,
+      "eval_steps_per_second": 0.836,
+      "step": 191
+    },
+    {
+      "epoch": 60.0,
+      "eval_accuracy": 0.3487544483985765,
+      "eval_loss": 1.711957573890686,
+      "eval_runtime": 3.0988,
+      "eval_samples_per_second": 90.681,
+      "eval_steps_per_second": 0.968,
+      "step": 195
+    },
+    {
+      "epoch": 60.92,
+      "eval_accuracy": 0.35587188612099646,
+      "eval_loss": 1.7048858404159546,
+      "eval_runtime": 3.3244,
+      "eval_samples_per_second": 84.526,
+      "eval_steps_per_second": 0.902,
+      "step": 198
+    },
+    {
+      "epoch": 61.54,
+      "grad_norm": 26972.24609375,
+      "learning_rate": 2.5925925925925925e-05,
+      "loss": 1.3094,
+      "step": 200
+    },
+    {
+      "epoch": 61.85,
+      "eval_accuracy": 0.3594306049822064,
+      "eval_loss": 1.702221155166626,
+      "eval_runtime": 2.9103,
+      "eval_samples_per_second": 96.553,
+      "eval_steps_per_second": 1.031,
+      "step": 201
+    },
+    {
+      "epoch": 62.77,
+      "eval_accuracy": 0.3736654804270463,
+      "eval_loss": 1.6912201642990112,
+      "eval_runtime": 3.4935,
+      "eval_samples_per_second": 80.435,
+      "eval_steps_per_second": 0.859,
+      "step": 204
+    },
+    {
+      "epoch": 64.0,
+      "eval_accuracy": 0.37722419928825623,
+      "eval_loss": 1.6797984838485718,
+      "eval_runtime": 3.0757,
+      "eval_samples_per_second": 91.361,
+      "eval_steps_per_second": 0.975,
+      "step": 208
+    },
+    {
+      "epoch": 64.92,
+      "eval_accuracy": 0.3807829181494662,
+      "eval_loss": 1.6687328815460205,
+      "eval_runtime": 3.281,
+      "eval_samples_per_second": 85.645,
+      "eval_steps_per_second": 0.914,
+      "step": 211
+    },
+    {
+      "epoch": 65.85,
+      "eval_accuracy": 0.38434163701067614,
+      "eval_loss": 1.6568727493286133,
+      "eval_runtime": 3.0158,
+      "eval_samples_per_second": 93.174,
+      "eval_steps_per_second": 0.995,
+      "step": 214
+    },
+    {
+      "epoch": 66.77,
+      "eval_accuracy": 0.3914590747330961,
+      "eval_loss": 1.642698049545288,
+      "eval_runtime": 2.9377,
+      "eval_samples_per_second": 95.654,
+      "eval_steps_per_second": 1.021,
+      "step": 217
+    },
+    {
+      "epoch": 68.0,
+      "eval_accuracy": 0.3914590747330961,
+      "eval_loss": 1.6301021575927734,
+      "eval_runtime": 2.9188,
+      "eval_samples_per_second": 96.272,
+      "eval_steps_per_second": 1.028,
+      "step": 221
+    },
+    {
+      "epoch": 68.92,
+      "eval_accuracy": 0.39501779359430605,
+      "eval_loss": 1.6217372417449951,
+      "eval_runtime": 3.1297,
+      "eval_samples_per_second": 89.784,
+      "eval_steps_per_second": 0.959,
+      "step": 224
+    },
+    {
+      "epoch": 69.85,
+      "eval_accuracy": 0.39501779359430605,
+      "eval_loss": 1.6203086376190186,
+      "eval_runtime": 3.3261,
+      "eval_samples_per_second": 84.482,
+      "eval_steps_per_second": 0.902,
+      "step": 227
+    },
+    {
+      "epoch": 70.77,
+      "eval_accuracy": 0.39501779359430605,
+      "eval_loss": 1.6257439851760864,
+      "eval_runtime": 3.1941,
+      "eval_samples_per_second": 87.974,
+      "eval_steps_per_second": 0.939,
+      "step": 230
+    },
+    {
+      "epoch": 72.0,
+      "eval_accuracy": 0.40213523131672596,
+      "eval_loss": 1.6192444562911987,
+      "eval_runtime": 2.8716,
+      "eval_samples_per_second": 97.855,
+      "eval_steps_per_second": 1.045,
+      "step": 234
+    },
+    {
+      "epoch": 72.92,
+      "eval_accuracy": 0.4092526690391459,
+      "eval_loss": 1.6044347286224365,
+      "eval_runtime": 3.3231,
+      "eval_samples_per_second": 84.559,
+      "eval_steps_per_second": 0.903,
+      "step": 237
+    },
+    {
+      "epoch": 73.85,
+      "eval_accuracy": 0.4306049822064057,
+      "eval_loss": 1.5868154764175415,
+      "eval_runtime": 3.0078,
+      "eval_samples_per_second": 93.422,
+      "eval_steps_per_second": 0.997,
+      "step": 240
+    },
+    {
+      "epoch": 74.77,
+      "eval_accuracy": 0.4377224199288256,
+      "eval_loss": 1.5786783695220947,
+      "eval_runtime": 3.1108,
+      "eval_samples_per_second": 90.332,
+      "eval_steps_per_second": 0.964,
+      "step": 243
+    },
+    {
+      "epoch": 76.0,
+      "eval_accuracy": 0.43416370106761565,
+      "eval_loss": 1.5762073993682861,
+      "eval_runtime": 4.8033,
+      "eval_samples_per_second": 58.501,
+      "eval_steps_per_second": 0.625,
+      "step": 247
+    },
+    {
+      "epoch": 76.92,
+      "eval_accuracy": 0.4377224199288256,
+      "eval_loss": 1.5717052221298218,
+      "eval_runtime": 4.9388,
+      "eval_samples_per_second": 56.896,
+      "eval_steps_per_second": 0.607,
+      "step": 250
+    },
+    {
+      "epoch": 77.85,
+      "eval_accuracy": 0.43416370106761565,
+      "eval_loss": 1.5673516988754272,
+      "eval_runtime": 3.5439,
+      "eval_samples_per_second": 79.29,
+      "eval_steps_per_second": 0.847,
+      "step": 253
+    },
+    {
+      "epoch": 78.77,
+      "eval_accuracy": 0.42704626334519574,
+      "eval_loss": 1.5683715343475342,
+      "eval_runtime": 2.9479,
+      "eval_samples_per_second": 95.323,
+      "eval_steps_per_second": 1.018,
+      "step": 256
+    },
+    {
+      "epoch": 80.0,
+      "eval_accuracy": 0.42704626334519574,
+      "eval_loss": 1.5619009733200073,
+      "eval_runtime": 3.2494,
+      "eval_samples_per_second": 86.478,
+      "eval_steps_per_second": 0.923,
+      "step": 260
+    },
+    {
+      "epoch": 80.92,
+      "eval_accuracy": 0.4306049822064057,
+      "eval_loss": 1.5554527044296265,
+      "eval_runtime": 3.0649,
+      "eval_samples_per_second": 91.683,
+      "eval_steps_per_second": 0.979,
+      "step": 263
+    },
+    {
+      "epoch": 81.85,
+      "eval_accuracy": 0.43416370106761565,
+      "eval_loss": 1.550489068031311,
+      "eval_runtime": 3.1587,
+      "eval_samples_per_second": 88.96,
+      "eval_steps_per_second": 0.95,
+      "step": 266
+    },
+    {
+      "epoch": 82.77,
+      "eval_accuracy": 0.4412811387900356,
+      "eval_loss": 1.5385645627975464,
+      "eval_runtime": 3.1715,
+      "eval_samples_per_second": 88.601,
+      "eval_steps_per_second": 0.946,
+      "step": 269
+    },
+    {
+      "epoch": 84.0,
+      "eval_accuracy": 0.4377224199288256,
+      "eval_loss": 1.536201000213623,
+      "eval_runtime": 3.2602,
+      "eval_samples_per_second": 86.191,
+      "eval_steps_per_second": 0.92,
+      "step": 273
+    },
+    {
+      "epoch": 84.92,
+      "eval_accuracy": 0.43416370106761565,
+      "eval_loss": 1.5410619974136353,
+      "eval_runtime": 2.9845,
+      "eval_samples_per_second": 94.153,
+      "eval_steps_per_second": 1.005,
+      "step": 276
+    },
+    {
+      "epoch": 85.85,
+      "eval_accuracy": 0.43416370106761565,
+      "eval_loss": 1.5452691316604614,
+      "eval_runtime": 3.4013,
+      "eval_samples_per_second": 82.616,
+      "eval_steps_per_second": 0.882,
+      "step": 279
+    },
+    {
+      "epoch": 86.77,
+      "eval_accuracy": 0.42704626334519574,
+      "eval_loss": 1.5611252784729004,
+      "eval_runtime": 2.9135,
+      "eval_samples_per_second": 96.447,
+      "eval_steps_per_second": 1.03,
+      "step": 282
+    },
+    {
+      "epoch": 88.0,
+      "eval_accuracy": 0.4199288256227758,
+      "eval_loss": 1.5766078233718872,
+      "eval_runtime": 2.8634,
+      "eval_samples_per_second": 98.135,
+      "eval_steps_per_second": 1.048,
+      "step": 286
+    },
+    {
+      "epoch": 88.92,
+      "eval_accuracy": 0.4199288256227758,
+      "eval_loss": 1.5781065225601196,
+      "eval_runtime": 3.1014,
+      "eval_samples_per_second": 90.606,
+      "eval_steps_per_second": 0.967,
+      "step": 289
+    },
+    {
+      "epoch": 89.85,
+      "eval_accuracy": 0.4234875444839858,
+      "eval_loss": 1.5674538612365723,
+      "eval_runtime": 3.5418,
+      "eval_samples_per_second": 79.339,
+      "eval_steps_per_second": 0.847,
+      "step": 292
+    },
+    {
+      "epoch": 90.77,
+      "eval_accuracy": 0.42704626334519574,
+      "eval_loss": 1.558840036392212,
+      "eval_runtime": 4.5717,
+      "eval_samples_per_second": 61.464,
+      "eval_steps_per_second": 0.656,
+      "step": 295
+    },
+    {
+      "epoch": 92.0,
+      "eval_accuracy": 0.42704626334519574,
+      "eval_loss": 1.5495978593826294,
+      "eval_runtime": 2.971,
+      "eval_samples_per_second": 94.581,
+      "eval_steps_per_second": 1.01,
+      "step": 299
+    },
+    {
+      "epoch": 92.31,
+      "grad_norm": 27984.919921875,
+      "learning_rate": 2.222222222222222e-05,
+      "loss": 1.0538,
+      "step": 300
+    },
+    {
+      "epoch": 92.92,
+      "eval_accuracy": 0.42704626334519574,
+      "eval_loss": 1.5492929220199585,
+      "eval_runtime": 3.229,
+      "eval_samples_per_second": 87.023,
+      "eval_steps_per_second": 0.929,
+      "step": 302
+    },
+    {
+      "epoch": 93.85,
+      "eval_accuracy": 0.4234875444839858,
+      "eval_loss": 1.5539740324020386,
+      "eval_runtime": 2.993,
+      "eval_samples_per_second": 93.886,
+      "eval_steps_per_second": 1.002,
+      "step": 305
+    },
+    {
+      "epoch": 94.77,
+      "eval_accuracy": 0.41637010676156583,
+      "eval_loss": 1.5620365142822266,
+      "eval_runtime": 3.5102,
+      "eval_samples_per_second": 80.052,
+      "eval_steps_per_second": 0.855,
+      "step": 308
+    },
+    {
+      "epoch": 96.0,
+      "eval_accuracy": 0.41637010676156583,
+      "eval_loss": 1.564751148223877,
+      "eval_runtime": 3.7132,
+      "eval_samples_per_second": 75.677,
+      "eval_steps_per_second": 0.808,
+      "step": 312
+    },
+    {
+      "epoch": 96.92,
+      "eval_accuracy": 0.41637010676156583,
+      "eval_loss": 1.561686396598816,
+      "eval_runtime": 4.9316,
+      "eval_samples_per_second": 56.98,
+      "eval_steps_per_second": 0.608,
+      "step": 315
+    },
+    {
+      "epoch": 97.85,
+      "eval_accuracy": 0.4234875444839858,
+      "eval_loss": 1.5461145639419556,
+      "eval_runtime": 3.1512,
+      "eval_samples_per_second": 89.173,
+      "eval_steps_per_second": 0.952,
+      "step": 318
+    },
+    {
+      "epoch": 98.77,
+      "eval_accuracy": 0.4306049822064057,
+      "eval_loss": 1.5348182916641235,
+      "eval_runtime": 4.3294,
+      "eval_samples_per_second": 64.906,
+      "eval_steps_per_second": 0.693,
+      "step": 321
+    },
+    {
+      "epoch": 100.0,
+      "eval_accuracy": 0.4306049822064057,
+      "eval_loss": 1.5345805883407593,
+      "eval_runtime": 3.3762,
+      "eval_samples_per_second": 83.23,
+      "eval_steps_per_second": 0.889,
+      "step": 325
+    },
+    {
+      "epoch": 100.92,
+      "eval_accuracy": 0.41637010676156583,
+      "eval_loss": 1.5465843677520752,
+      "eval_runtime": 3.8288,
+      "eval_samples_per_second": 73.391,
+      "eval_steps_per_second": 0.784,
+      "step": 328
+    },
+    {
+      "epoch": 101.85,
+      "eval_accuracy": 0.4128113879003559,
+      "eval_loss": 1.5547189712524414,
+      "eval_runtime": 4.3332,
+      "eval_samples_per_second": 64.848,
+      "eval_steps_per_second": 0.692,
+      "step": 331
+    },
+    {
+      "epoch": 102.77,
+      "eval_accuracy": 0.4128113879003559,
+      "eval_loss": 1.5559605360031128,
+      "eval_runtime": 3.2588,
+      "eval_samples_per_second": 86.229,
+      "eval_steps_per_second": 0.921,
+      "step": 334
+    },
+    {
+      "epoch": 104.0,
+      "eval_accuracy": 0.4306049822064057,
+      "eval_loss": 1.5315039157867432,
+      "eval_runtime": 4.5744,
+      "eval_samples_per_second": 61.429,
+      "eval_steps_per_second": 0.656,
+      "step": 338
+    },
+    {
+      "epoch": 104.92,
+      "eval_accuracy": 0.44483985765124556,
+      "eval_loss": 1.5124022960662842,
+      "eval_runtime": 3.3067,
+      "eval_samples_per_second": 84.979,
+      "eval_steps_per_second": 0.907,
+      "step": 341
+    },
+    {
+      "epoch": 105.85,
+      "eval_accuracy": 0.44483985765124556,
+      "eval_loss": 1.5044087171554565,
+      "eval_runtime": 3.9949,
+      "eval_samples_per_second": 70.341,
+      "eval_steps_per_second": 0.751,
+      "step": 344
+    },
+    {
+      "epoch": 106.77,
+      "eval_accuracy": 0.4483985765124555,
+      "eval_loss": 1.5010027885437012,
+      "eval_runtime": 3.5698,
+      "eval_samples_per_second": 78.716,
+      "eval_steps_per_second": 0.84,
+      "step": 347
+    },
+    {
+      "epoch": 108.0,
+      "eval_accuracy": 0.44483985765124556,
+      "eval_loss": 1.5004721879959106,
+      "eval_runtime": 2.9807,
+      "eval_samples_per_second": 94.273,
+      "eval_steps_per_second": 1.006,
+      "step": 351
+    },
+    {
+      "epoch": 108.92,
+      "eval_accuracy": 0.44483985765124556,
+      "eval_loss": 1.499153971672058,
+      "eval_runtime": 2.8868,
+      "eval_samples_per_second": 97.339,
+      "eval_steps_per_second": 1.039,
+      "step": 354
+    },
+    {
+      "epoch": 109.85,
+      "eval_accuracy": 0.4483985765124555,
+      "eval_loss": 1.4993938207626343,
+      "eval_runtime": 3.2052,
+      "eval_samples_per_second": 87.67,
+      "eval_steps_per_second": 0.936,
+      "step": 357
+    },
+    {
+      "epoch": 110.77,
+      "eval_accuracy": 0.45195729537366547,
+      "eval_loss": 1.4987653493881226,
+      "eval_runtime": 3.3473,
+      "eval_samples_per_second": 83.949,
+      "eval_steps_per_second": 0.896,
+      "step": 360
+    },
+    {
+      "epoch": 112.0,
+      "eval_accuracy": 0.46619217081850534,
+      "eval_loss": 1.5004514455795288,
+      "eval_runtime": 2.8714,
+      "eval_samples_per_second": 97.862,
+      "eval_steps_per_second": 1.045,
+      "step": 364
+    },
+    {
+      "epoch": 112.92,
+      "eval_accuracy": 0.47330960854092524,
+      "eval_loss": 1.5010361671447754,
+      "eval_runtime": 3.6886,
+      "eval_samples_per_second": 76.182,
+      "eval_steps_per_second": 0.813,
+      "step": 367
+    },
+    {
+      "epoch": 113.85,
+      "eval_accuracy": 0.4697508896797153,
+      "eval_loss": 1.4968541860580444,
+      "eval_runtime": 3.5621,
+      "eval_samples_per_second": 78.886,
+      "eval_steps_per_second": 0.842,
+      "step": 370
+    },
+    {
+      "epoch": 114.77,
+      "eval_accuracy": 0.47330960854092524,
+      "eval_loss": 1.4775702953338623,
+      "eval_runtime": 4.3842,
+      "eval_samples_per_second": 64.093,
+      "eval_steps_per_second": 0.684,
+      "step": 373
+    },
+    {
+      "epoch": 116.0,
+      "eval_accuracy": 0.47686832740213525,
+      "eval_loss": 1.4527899026870728,
+      "eval_runtime": 4.7808,
+      "eval_samples_per_second": 58.777,
+      "eval_steps_per_second": 0.628,
+      "step": 377
+    },
+    {
+      "epoch": 116.92,
+      "eval_accuracy": 0.49466192170818507,
+      "eval_loss": 1.4394866228103638,
+      "eval_runtime": 5.0753,
+      "eval_samples_per_second": 55.366,
+      "eval_steps_per_second": 0.591,
+      "step": 380
+    },
+    {
+      "epoch": 117.85,
+      "eval_accuracy": 0.498220640569395,
+      "eval_loss": 1.4310173988342285,
+      "eval_runtime": 4.758,
+      "eval_samples_per_second": 59.058,
+      "eval_steps_per_second": 0.631,
+      "step": 383
+    },
+    {
+      "epoch": 118.77,
+      "eval_accuracy": 0.49466192170818507,
+      "eval_loss": 1.4314603805541992,
+      "eval_runtime": 3.9673,
+      "eval_samples_per_second": 70.829,
+      "eval_steps_per_second": 0.756,
+      "step": 386
+    },
+    {
+      "epoch": 120.0,
+      "eval_accuracy": 0.49466192170818507,
+      "eval_loss": 1.4388599395751953,
+      "eval_runtime": 4.1069,
+      "eval_samples_per_second": 68.422,
+      "eval_steps_per_second": 0.73,
+      "step": 390
+    },
+    {
+      "epoch": 120.92,
+      "eval_accuracy": 0.498220640569395,
+      "eval_loss": 1.4374699592590332,
+      "eval_runtime": 5.1154,
+      "eval_samples_per_second": 54.933,
+      "eval_steps_per_second": 0.586,
+      "step": 393
+    },
+    {
+      "epoch": 121.85,
+      "eval_accuracy": 0.498220640569395,
+      "eval_loss": 1.4381343126296997,
+      "eval_runtime": 4.1133,
+      "eval_samples_per_second": 68.315,
+      "eval_steps_per_second": 0.729,
+      "step": 396
+    },
+    {
+      "epoch": 122.77,
+      "eval_accuracy": 0.498220640569395,
+      "eval_loss": 1.4246776103973389,
+      "eval_runtime": 3.9833,
+      "eval_samples_per_second": 70.544,
+      "eval_steps_per_second": 0.753,
+      "step": 399
+    },
+    {
+      "epoch": 123.08,
+      "grad_norm": 31388.482421875,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.8509,
+      "step": 400
+    },
+    {
+      "epoch": 124.0,
+      "eval_accuracy": 0.498220640569395,
+      "eval_loss": 1.4195659160614014,
+      "eval_runtime": 4.1654,
+      "eval_samples_per_second": 67.461,
+      "eval_steps_per_second": 0.72,
+      "step": 403
+    },
+    {
+      "epoch": 124.92,
+      "eval_accuracy": 0.505338078291815,
+      "eval_loss": 1.4178649187088013,
+      "eval_runtime": 5.0869,
+      "eval_samples_per_second": 55.239,
+      "eval_steps_per_second": 0.59,
+      "step": 406
+    },
+    {
+      "epoch": 125.85,
+      "eval_accuracy": 0.505338078291815,
+      "eval_loss": 1.40910804271698,
+      "eval_runtime": 4.5242,
+      "eval_samples_per_second": 62.11,
+      "eval_steps_per_second": 0.663,
+      "step": 409
+    },
+    {
+      "epoch": 126.77,
+      "eval_accuracy": 0.505338078291815,
+      "eval_loss": 1.3957635164260864,
+      "eval_runtime": 4.5377,
+      "eval_samples_per_second": 61.926,
+      "eval_steps_per_second": 0.661,
+      "step": 412
+    },
+    {
+      "epoch": 128.0,
+      "eval_accuracy": 0.5088967971530249,
+      "eval_loss": 1.3736003637313843,
+      "eval_runtime": 3.6994,
+      "eval_samples_per_second": 75.958,
+      "eval_steps_per_second": 0.811,
+      "step": 416
+    },
+    {
+      "epoch": 128.92,
+      "eval_accuracy": 0.5088967971530249,
+      "eval_loss": 1.3661431074142456,
+      "eval_runtime": 4.0248,
+      "eval_samples_per_second": 69.817,
+      "eval_steps_per_second": 0.745,
+      "step": 419
+    },
+    {
+      "epoch": 129.85,
+      "eval_accuracy": 0.5124555160142349,
+      "eval_loss": 1.369443416595459,
+      "eval_runtime": 4.9876,
+      "eval_samples_per_second": 56.34,
+      "eval_steps_per_second": 0.601,
+      "step": 422
+    },
+    {
+      "epoch": 130.77,
+      "eval_accuracy": 0.5124555160142349,
+      "eval_loss": 1.3807623386383057,
+      "eval_runtime": 3.5494,
+      "eval_samples_per_second": 79.169,
+      "eval_steps_per_second": 0.845,
+      "step": 425
+    },
+    {
+      "epoch": 132.0,
+      "eval_accuracy": 0.5124555160142349,
+      "eval_loss": 1.3818711042404175,
+      "eval_runtime": 3.9503,
+      "eval_samples_per_second": 71.134,
+      "eval_steps_per_second": 0.759,
+      "step": 429
+    },
+    {
+      "epoch": 132.92,
+      "eval_accuracy": 0.5124555160142349,
+      "eval_loss": 1.3859163522720337,
+      "eval_runtime": 4.2041,
+      "eval_samples_per_second": 66.84,
+      "eval_steps_per_second": 0.714,
+      "step": 432
+    },
+    {
+      "epoch": 133.85,
+      "eval_accuracy": 0.5231316725978647,
+      "eval_loss": 1.378004789352417,
+      "eval_runtime": 3.8384,
+      "eval_samples_per_second": 73.208,
+      "eval_steps_per_second": 0.782,
+      "step": 435
+    },
+    {
+      "epoch": 134.77,
+      "eval_accuracy": 0.5231316725978647,
+      "eval_loss": 1.3696413040161133,
+      "eval_runtime": 4.6334,
+      "eval_samples_per_second": 60.646,
+      "eval_steps_per_second": 0.647,
+      "step": 438
+    },
+    {
+      "epoch": 136.0,
+      "eval_accuracy": 0.5302491103202847,
+      "eval_loss": 1.3564013242721558,
+      "eval_runtime": 4.002,
+      "eval_samples_per_second": 70.215,
+      "eval_steps_per_second": 0.75,
+      "step": 442
+    },
+    {
+      "epoch": 136.92,
+      "eval_accuracy": 0.5338078291814946,
+      "eval_loss": 1.3421210050582886,
+      "eval_runtime": 4.0161,
+      "eval_samples_per_second": 69.968,
+      "eval_steps_per_second": 0.747,
+      "step": 445
+    },
+    {
+      "epoch": 137.85,
+      "eval_accuracy": 0.5373665480427047,
+      "eval_loss": 1.325627326965332,
+      "eval_runtime": 4.156,
+      "eval_samples_per_second": 67.613,
+      "eval_steps_per_second": 0.722,
+      "step": 448
+    },
+    {
+      "epoch": 138.77,
+      "eval_accuracy": 0.5373665480427047,
+      "eval_loss": 1.3274290561676025,
+      "eval_runtime": 3.9911,
+      "eval_samples_per_second": 70.407,
+      "eval_steps_per_second": 0.752,
+      "step": 451
+    },
+    {
+      "epoch": 140.0,
+      "eval_accuracy": 0.5409252669039146,
+      "eval_loss": 1.3401566743850708,
+      "eval_runtime": 4.4088,
+      "eval_samples_per_second": 63.736,
+      "eval_steps_per_second": 0.68,
+      "step": 455
+    },
+    {
+      "epoch": 140.92,
+      "eval_accuracy": 0.5409252669039146,
+      "eval_loss": 1.351689338684082,
+      "eval_runtime": 4.4409,
+      "eval_samples_per_second": 63.276,
+      "eval_steps_per_second": 0.676,
+      "step": 458
+    },
+    {
+      "epoch": 141.85,
+      "eval_accuracy": 0.5409252669039146,
+      "eval_loss": 1.3585495948791504,
+      "eval_runtime": 3.7955,
+      "eval_samples_per_second": 74.035,
+      "eval_steps_per_second": 0.79,
+      "step": 461
+    },
+    {
+      "epoch": 142.77,
+      "eval_accuracy": 0.5373665480427047,
+      "eval_loss": 1.3592112064361572,
+      "eval_runtime": 3.3552,
+      "eval_samples_per_second": 83.75,
+      "eval_steps_per_second": 0.894,
+      "step": 464
+    },
+    {
+      "epoch": 144.0,
+      "eval_accuracy": 0.5480427046263345,
+      "eval_loss": 1.3329293727874756,
+      "eval_runtime": 5.3044,
+      "eval_samples_per_second": 52.975,
+      "eval_steps_per_second": 0.566,
+      "step": 468
+    },
+    {
+      "epoch": 144.92,
+      "eval_accuracy": 0.5480427046263345,
+      "eval_loss": 1.312560796737671,
+      "eval_runtime": 4.319,
+      "eval_samples_per_second": 65.061,
+      "eval_steps_per_second": 0.695,
+      "step": 471
+    },
+    {
+      "epoch": 145.85,
+      "eval_accuracy": 0.5444839857651246,
+      "eval_loss": 1.3075566291809082,
+      "eval_runtime": 3.9528,
+      "eval_samples_per_second": 71.09,
+      "eval_steps_per_second": 0.759,
+      "step": 474
+    },
+    {
+      "epoch": 146.77,
+      "eval_accuracy": 0.5480427046263345,
+      "eval_loss": 1.3146412372589111,
+      "eval_runtime": 4.3249,
+      "eval_samples_per_second": 64.973,
+      "eval_steps_per_second": 0.694,
+      "step": 477
+    },
+    {
+      "epoch": 148.0,
+      "eval_accuracy": 0.5444839857651246,
+      "eval_loss": 1.3345069885253906,
+      "eval_runtime": 3.9127,
+      "eval_samples_per_second": 71.817,
+      "eval_steps_per_second": 0.767,
+      "step": 481
+    },
+    {
+      "epoch": 148.92,
+      "eval_accuracy": 0.5444839857651246,
+      "eval_loss": 1.3408929109573364,
+      "eval_runtime": 4.1463,
+      "eval_samples_per_second": 67.771,
+      "eval_steps_per_second": 0.724,
+      "step": 484
+    },
+    {
+      "epoch": 149.85,
+      "eval_accuracy": 0.5444839857651246,
+      "eval_loss": 1.3374032974243164,
+      "eval_runtime": 4.2345,
+      "eval_samples_per_second": 66.359,
+      "eval_steps_per_second": 0.708,
+      "step": 487
+    },
+    {
+      "epoch": 150.77,
+      "eval_accuracy": 0.5480427046263345,
+      "eval_loss": 1.3227189779281616,
+      "eval_runtime": 4.3006,
+      "eval_samples_per_second": 65.339,
+      "eval_steps_per_second": 0.698,
+      "step": 490
+    },
+    {
+      "epoch": 152.0,
+      "eval_accuracy": 0.5444839857651246,
+      "eval_loss": 1.3200651407241821,
+      "eval_runtime": 4.4216,
+      "eval_samples_per_second": 63.551,
+      "eval_steps_per_second": 0.678,
+      "step": 494
+    },
+    {
+      "epoch": 152.92,
+      "eval_accuracy": 0.5444839857651246,
+      "eval_loss": 1.3174102306365967,
+      "eval_runtime": 4.4898,
+      "eval_samples_per_second": 62.586,
+      "eval_steps_per_second": 0.668,
+      "step": 497
+    },
+    {
+      "epoch": 153.85,
+      "grad_norm": 24984.7734375,
+      "learning_rate": 1.4814814814814815e-05,
+      "loss": 0.7118,
+      "step": 500
+    },
+    {
+      "epoch": 153.85,
+      "eval_accuracy": 0.5444839857651246,
+      "eval_loss": 1.3073471784591675,
+      "eval_runtime": 4.2385,
+      "eval_samples_per_second": 66.297,
+      "eval_steps_per_second": 0.708,
+      "step": 500
+    },
+    {
+      "epoch": 154.77,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.2983657121658325,
+      "eval_runtime": 3.4569,
+      "eval_samples_per_second": 81.286,
+      "eval_steps_per_second": 0.868,
+      "step": 503
+    },
+    {
+      "epoch": 156.0,
+      "eval_accuracy": 0.5516014234875445,
+      "eval_loss": 1.2974605560302734,
+      "eval_runtime": 4.3467,
+      "eval_samples_per_second": 64.647,
+      "eval_steps_per_second": 0.69,
+      "step": 507
+    },
+    {
+      "epoch": 156.92,
+      "eval_accuracy": 0.5516014234875445,
+      "eval_loss": 1.3027478456497192,
+      "eval_runtime": 4.5106,
+      "eval_samples_per_second": 62.297,
+      "eval_steps_per_second": 0.665,
+      "step": 510
+    },
+    {
+      "epoch": 157.85,
+      "eval_accuracy": 0.5480427046263345,
+      "eval_loss": 1.3088507652282715,
+      "eval_runtime": 4.2508,
+      "eval_samples_per_second": 66.105,
+      "eval_steps_per_second": 0.706,
+      "step": 513
+    },
+    {
+      "epoch": 158.77,
+      "eval_accuracy": 0.5480427046263345,
+      "eval_loss": 1.3138750791549683,
+      "eval_runtime": 4.4205,
+      "eval_samples_per_second": 63.567,
+      "eval_steps_per_second": 0.679,
+      "step": 516
+    },
+    {
+      "epoch": 160.0,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.3067928552627563,
+      "eval_runtime": 4.2488,
+      "eval_samples_per_second": 66.136,
+      "eval_steps_per_second": 0.706,
+      "step": 520
+    },
+    {
+      "epoch": 160.92,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.3011025190353394,
+      "eval_runtime": 4.3025,
+      "eval_samples_per_second": 65.31,
+      "eval_steps_per_second": 0.697,
+      "step": 523
+    },
+    {
+      "epoch": 161.85,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.2957364320755005,
+      "eval_runtime": 4.3812,
+      "eval_samples_per_second": 64.137,
+      "eval_steps_per_second": 0.685,
+      "step": 526
+    },
+    {
+      "epoch": 162.77,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.296021819114685,
+      "eval_runtime": 4.6921,
+      "eval_samples_per_second": 59.887,
+      "eval_steps_per_second": 0.639,
+      "step": 529
+    },
+    {
+      "epoch": 164.0,
+      "eval_accuracy": 0.5516014234875445,
+      "eval_loss": 1.3158953189849854,
+      "eval_runtime": 4.7452,
+      "eval_samples_per_second": 59.218,
+      "eval_steps_per_second": 0.632,
+      "step": 533
+    },
+    {
+      "epoch": 164.92,
+      "eval_accuracy": 0.5516014234875445,
+      "eval_loss": 1.3257168531417847,
+      "eval_runtime": 4.4128,
+      "eval_samples_per_second": 63.678,
+      "eval_steps_per_second": 0.68,
+      "step": 536
+    },
+    {
+      "epoch": 165.85,
+      "eval_accuracy": 0.5516014234875445,
+      "eval_loss": 1.3312301635742188,
+      "eval_runtime": 3.6447,
+      "eval_samples_per_second": 77.099,
+      "eval_steps_per_second": 0.823,
+      "step": 539
+    },
+    {
+      "epoch": 166.77,
+      "eval_accuracy": 0.5516014234875445,
+      "eval_loss": 1.322218418121338,
+      "eval_runtime": 4.2773,
+      "eval_samples_per_second": 65.695,
+      "eval_steps_per_second": 0.701,
+      "step": 542
+    },
+    {
+      "epoch": 168.0,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.298622488975525,
+      "eval_runtime": 4.5789,
+      "eval_samples_per_second": 61.369,
+      "eval_steps_per_second": 0.655,
+      "step": 546
+    },
+    {
+      "epoch": 168.92,
+      "eval_accuracy": 0.5587188612099644,
+      "eval_loss": 1.289797306060791,
+      "eval_runtime": 4.5328,
+      "eval_samples_per_second": 61.993,
+      "eval_steps_per_second": 0.662,
+      "step": 549
+    },
+    {
+      "epoch": 169.85,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.2937852144241333,
+      "eval_runtime": 3.7509,
+      "eval_samples_per_second": 74.915,
+      "eval_steps_per_second": 0.8,
+      "step": 552
+    },
+    {
+      "epoch": 170.77,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.290231704711914,
+      "eval_runtime": 4.1153,
+      "eval_samples_per_second": 68.282,
+      "eval_steps_per_second": 0.729,
+      "step": 555
+    },
+    {
+      "epoch": 172.0,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.287913203239441,
+      "eval_runtime": 4.7912,
+      "eval_samples_per_second": 58.649,
+      "eval_steps_per_second": 0.626,
+      "step": 559
+    },
+    {
+      "epoch": 172.92,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.283803939819336,
+      "eval_runtime": 4.5456,
+      "eval_samples_per_second": 61.818,
+      "eval_steps_per_second": 0.66,
+      "step": 562
+    },
+    {
+      "epoch": 173.85,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2811965942382812,
+      "eval_runtime": 4.4869,
+      "eval_samples_per_second": 62.627,
+      "eval_steps_per_second": 0.669,
+      "step": 565
+    },
+    {
+      "epoch": 174.77,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2863661050796509,
+      "eval_runtime": 4.2715,
+      "eval_samples_per_second": 65.785,
+      "eval_steps_per_second": 0.702,
+      "step": 568
+    },
+    {
+      "epoch": 176.0,
+      "eval_accuracy": 0.5551601423487544,
+      "eval_loss": 1.2934131622314453,
+      "eval_runtime": 4.643,
+      "eval_samples_per_second": 60.522,
+      "eval_steps_per_second": 0.646,
+      "step": 572
+    },
+    {
+      "epoch": 176.92,
+      "eval_accuracy": 0.5587188612099644,
+      "eval_loss": 1.2940202951431274,
+      "eval_runtime": 4.2681,
+      "eval_samples_per_second": 65.837,
+      "eval_steps_per_second": 0.703,
+      "step": 575
+    },
+    {
+      "epoch": 177.85,
+      "eval_accuracy": 0.5587188612099644,
+      "eval_loss": 1.298832654953003,
+      "eval_runtime": 4.2991,
+      "eval_samples_per_second": 65.363,
+      "eval_steps_per_second": 0.698,
+      "step": 578
+    },
+    {
+      "epoch": 178.77,
+      "eval_accuracy": 0.5622775800711743,
+      "eval_loss": 1.295286774635315,
+      "eval_runtime": 4.1989,
+      "eval_samples_per_second": 66.922,
+      "eval_steps_per_second": 0.714,
+      "step": 581
+    },
+    {
+      "epoch": 180.0,
+      "eval_accuracy": 0.5587188612099644,
+      "eval_loss": 1.2971975803375244,
+      "eval_runtime": 4.7188,
+      "eval_samples_per_second": 59.549,
+      "eval_steps_per_second": 0.636,
+      "step": 585
+    },
+    {
+      "epoch": 180.92,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2936004400253296,
+      "eval_runtime": 4.813,
+      "eval_samples_per_second": 58.383,
+      "eval_steps_per_second": 0.623,
+      "step": 588
+    },
+    {
+      "epoch": 181.85,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2928047180175781,
+      "eval_runtime": 4.1735,
+      "eval_samples_per_second": 67.33,
+      "eval_steps_per_second": 0.719,
+      "step": 591
+    },
+    {
+      "epoch": 182.77,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.291295051574707,
+      "eval_runtime": 4.4694,
+      "eval_samples_per_second": 62.872,
+      "eval_steps_per_second": 0.671,
+      "step": 594
+    },
+    {
+      "epoch": 184.0,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2824889421463013,
+      "eval_runtime": 4.0765,
+      "eval_samples_per_second": 68.932,
+      "eval_steps_per_second": 0.736,
+      "step": 598
+    },
+    {
+      "epoch": 184.62,
+      "grad_norm": 29892.71484375,
+      "learning_rate": 1.111111111111111e-05,
+      "loss": 0.6473,
+      "step": 600
+    },
+    {
+      "epoch": 184.92,
+      "eval_accuracy": 0.5693950177935944,
+      "eval_loss": 1.2735832929611206,
+      "eval_runtime": 4.6704,
+      "eval_samples_per_second": 60.166,
+      "eval_steps_per_second": 0.642,
+      "step": 601
+    },
+    {
+      "epoch": 185.85,
+      "eval_accuracy": 0.5693950177935944,
+      "eval_loss": 1.2714898586273193,
+      "eval_runtime": 4.6432,
+      "eval_samples_per_second": 60.519,
+      "eval_steps_per_second": 0.646,
+      "step": 604
+    },
+    {
+      "epoch": 186.77,
+      "eval_accuracy": 0.5693950177935944,
+      "eval_loss": 1.2703534364700317,
+      "eval_runtime": 4.1853,
+      "eval_samples_per_second": 67.139,
+      "eval_steps_per_second": 0.717,
+      "step": 607
+    },
+    {
+      "epoch": 188.0,
+      "eval_accuracy": 0.5693950177935944,
+      "eval_loss": 1.2716755867004395,
+      "eval_runtime": 4.1775,
+      "eval_samples_per_second": 67.265,
+      "eval_steps_per_second": 0.718,
+      "step": 611
+    },
+    {
+      "epoch": 188.92,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2724348306655884,
+      "eval_runtime": 4.9312,
+      "eval_samples_per_second": 56.984,
+      "eval_steps_per_second": 0.608,
+      "step": 614
+    },
+    {
+      "epoch": 189.85,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2763242721557617,
+      "eval_runtime": 3.5712,
+      "eval_samples_per_second": 78.685,
+      "eval_steps_per_second": 0.84,
+      "step": 617
+    },
+    {
+      "epoch": 190.77,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2811599969863892,
+      "eval_runtime": 4.2324,
+      "eval_samples_per_second": 66.393,
+      "eval_steps_per_second": 0.709,
+      "step": 620
+    },
+    {
+      "epoch": 192.0,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2791301012039185,
+      "eval_runtime": 4.4625,
+      "eval_samples_per_second": 62.97,
+      "eval_steps_per_second": 0.672,
+      "step": 624
+    },
+    {
+      "epoch": 192.92,
+      "eval_accuracy": 0.5693950177935944,
+      "eval_loss": 1.2697654962539673,
+      "eval_runtime": 4.2766,
+      "eval_samples_per_second": 65.707,
+      "eval_steps_per_second": 0.701,
+      "step": 627
+    },
+    {
+      "epoch": 193.85,
+      "eval_accuracy": 0.5693950177935944,
+      "eval_loss": 1.269476294517517,
+      "eval_runtime": 4.2862,
+      "eval_samples_per_second": 65.56,
+      "eval_steps_per_second": 0.7,
+      "step": 630
+    },
+    {
+      "epoch": 194.77,
+      "eval_accuracy": 0.5693950177935944,
+      "eval_loss": 1.2703962326049805,
+      "eval_runtime": 4.2135,
+      "eval_samples_per_second": 66.69,
+      "eval_steps_per_second": 0.712,
+      "step": 633
+    },
+    {
+      "epoch": 196.0,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2736749649047852,
+      "eval_runtime": 4.1666,
+      "eval_samples_per_second": 67.441,
+      "eval_steps_per_second": 0.72,
+      "step": 637
+    },
+    {
+      "epoch": 196.92,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2782082557678223,
+      "eval_runtime": 4.3682,
+      "eval_samples_per_second": 64.329,
+      "eval_steps_per_second": 0.687,
+      "step": 640
+    },
+    {
+      "epoch": 197.85,
+      "eval_accuracy": 0.5622775800711743,
+      "eval_loss": 1.2813825607299805,
+      "eval_runtime": 5.6488,
+      "eval_samples_per_second": 49.745,
+      "eval_steps_per_second": 0.531,
+      "step": 643
+    },
+    {
+      "epoch": 198.77,
+      "eval_accuracy": 0.5622775800711743,
+      "eval_loss": 1.2819089889526367,
+      "eval_runtime": 5.1916,
+      "eval_samples_per_second": 54.126,
+      "eval_steps_per_second": 0.578,
+      "step": 646
+    },
+    {
+      "epoch": 200.0,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.274595022201538,
+      "eval_runtime": 4.4378,
+      "eval_samples_per_second": 63.32,
+      "eval_steps_per_second": 0.676,
+      "step": 650
+    },
+    {
+      "epoch": 200.92,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2694467306137085,
+      "eval_runtime": 4.797,
+      "eval_samples_per_second": 58.579,
+      "eval_steps_per_second": 0.625,
+      "step": 653
+    },
+    {
+      "epoch": 201.85,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.262547254562378,
+      "eval_runtime": 4.6991,
+      "eval_samples_per_second": 59.798,
+      "eval_steps_per_second": 0.638,
+      "step": 656
+    },
+    {
+      "epoch": 202.77,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2575123310089111,
+      "eval_runtime": 4.9663,
+      "eval_samples_per_second": 56.582,
+      "eval_steps_per_second": 0.604,
+      "step": 659
+    },
+    {
+      "epoch": 204.0,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2548755407333374,
+      "eval_runtime": 5.2012,
+      "eval_samples_per_second": 54.026,
+      "eval_steps_per_second": 0.577,
+      "step": 663
+    },
+    {
+      "epoch": 204.92,
+      "eval_accuracy": 0.5729537366548043,
+      "eval_loss": 1.2623133659362793,
+      "eval_runtime": 4.5347,
+      "eval_samples_per_second": 61.967,
+      "eval_steps_per_second": 0.662,
+      "step": 666
+    },
+    {
+      "epoch": 205.85,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2665455341339111,
+      "eval_runtime": 3.1603,
+      "eval_samples_per_second": 88.917,
+      "eval_steps_per_second": 0.949,
+      "step": 669
+    },
+    {
+      "epoch": 206.77,
+      "eval_accuracy": 0.5658362989323843,
+      "eval_loss": 1.2684026956558228,
+      "eval_runtime": 4.2009,
+      "eval_samples_per_second": 66.89,
+      "eval_steps_per_second": 0.714,
+      "step": 672
+    },
+    {
+      "epoch": 208.0,
+      "eval_accuracy": 0.5622775800711743,
+      "eval_loss": 1.277047038078308,
+      "eval_runtime": 4.3489,
+      "eval_samples_per_second": 64.613,
+      "eval_steps_per_second": 0.69,
+      "step": 676
+    },
+    {
+      "epoch": 208.92,
+      "eval_accuracy": 0.5622775800711743,
+      "eval_loss": 1.2807551622390747,
+      "eval_runtime": 3.8563,
+      "eval_samples_per_second": 72.867,
+      "eval_steps_per_second": 0.778,
+      "step": 679
+    },
+    {
+      "epoch": 209.85,
+      "eval_accuracy": 0.5729537366548043,
+      "eval_loss": 1.2761532068252563,
+      "eval_runtime": 4.7161,
+      "eval_samples_per_second": 59.583,
+      "eval_steps_per_second": 0.636,
+      "step": 682
+    },
+    {
+      "epoch": 210.77,
+      "eval_accuracy": 0.5729537366548043,
+      "eval_loss": 1.2759194374084473,
+      "eval_runtime": 5.0376,
+      "eval_samples_per_second": 55.781,
+      "eval_steps_per_second": 0.596,
+      "step": 685
+    },
+    {
+      "epoch": 212.0,
+      "eval_accuracy": 0.5729537366548043,
+      "eval_loss": 1.2752187252044678,
+      "eval_runtime": 4.5842,
+      "eval_samples_per_second": 61.297,
+      "eval_steps_per_second": 0.654,
+      "step": 689
+    },
+    {
+      "epoch": 212.92,
+      "eval_accuracy": 0.5729537366548043,
+      "eval_loss": 1.275394082069397,
+      "eval_runtime": 4.2209,
+      "eval_samples_per_second": 66.573,
+      "eval_steps_per_second": 0.711,
+      "step": 692
+    },
+    {
+      "epoch": 213.85,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.272161602973938,
+      "eval_runtime": 4.7348,
+      "eval_samples_per_second": 59.347,
+      "eval_steps_per_second": 0.634,
+      "step": 695
+    },
+    {
+      "epoch": 214.77,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.273858904838562,
+      "eval_runtime": 4.0254,
+      "eval_samples_per_second": 69.808,
+      "eval_steps_per_second": 0.745,
+      "step": 698
+    },
+    {
+      "epoch": 215.38,
+      "grad_norm": 28098.056640625,
+      "learning_rate": 7.4074074074074075e-06,
+      "loss": 0.613,
+      "step": 700
+    },
+    {
+      "epoch": 216.0,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2782981395721436,
+      "eval_runtime": 4.6723,
+      "eval_samples_per_second": 60.142,
+      "eval_steps_per_second": 0.642,
+      "step": 702
+    },
+    {
+      "epoch": 216.92,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2774933576583862,
+      "eval_runtime": 4.576,
+      "eval_samples_per_second": 61.407,
+      "eval_steps_per_second": 0.656,
+      "step": 705
+    },
+    {
+      "epoch": 217.85,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2740654945373535,
+      "eval_runtime": 4.8253,
+      "eval_samples_per_second": 58.234,
+      "eval_steps_per_second": 0.622,
+      "step": 708
+    },
+    {
+      "epoch": 218.77,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2705509662628174,
+      "eval_runtime": 4.386,
+      "eval_samples_per_second": 64.067,
+      "eval_steps_per_second": 0.684,
+      "step": 711
+    },
+    {
+      "epoch": 220.0,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2627956867218018,
+      "eval_runtime": 4.2817,
+      "eval_samples_per_second": 65.628,
+      "eval_steps_per_second": 0.701,
+      "step": 715
+    },
+    {
+      "epoch": 220.92,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2580970525741577,
+      "eval_runtime": 3.9386,
+      "eval_samples_per_second": 71.344,
+      "eval_steps_per_second": 0.762,
+      "step": 718
+    },
+    {
+      "epoch": 221.85,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2567566633224487,
+      "eval_runtime": 4.3353,
+      "eval_samples_per_second": 64.817,
+      "eval_steps_per_second": 0.692,
+      "step": 721
+    },
+    {
+      "epoch": 222.77,
+      "eval_accuracy": 0.5729537366548043,
+      "eval_loss": 1.2558982372283936,
+      "eval_runtime": 3.7135,
+      "eval_samples_per_second": 75.67,
+      "eval_steps_per_second": 0.808,
+      "step": 724
+    },
+    {
+      "epoch": 224.0,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2502700090408325,
+      "eval_runtime": 5.0636,
+      "eval_samples_per_second": 55.494,
+      "eval_steps_per_second": 0.592,
+      "step": 728
+    },
+    {
+      "epoch": 224.92,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2497973442077637,
+      "eval_runtime": 4.7669,
+      "eval_samples_per_second": 58.948,
+      "eval_steps_per_second": 0.629,
+      "step": 731
+    },
+    {
+      "epoch": 225.85,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2500195503234863,
+      "eval_runtime": 3.9522,
+      "eval_samples_per_second": 71.099,
+      "eval_steps_per_second": 0.759,
+      "step": 734
+    },
+    {
+      "epoch": 226.77,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2490234375,
+      "eval_runtime": 4.1869,
+      "eval_samples_per_second": 67.114,
+      "eval_steps_per_second": 0.717,
+      "step": 737
+    },
+    {
+      "epoch": 228.0,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2531741857528687,
+      "eval_runtime": 3.9865,
+      "eval_samples_per_second": 70.489,
+      "eval_steps_per_second": 0.753,
+      "step": 741
+    },
+    {
+      "epoch": 228.92,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2572293281555176,
+      "eval_runtime": 5.2241,
+      "eval_samples_per_second": 53.789,
+      "eval_steps_per_second": 0.574,
+      "step": 744
+    },
+    {
+      "epoch": 229.85,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2598803043365479,
+      "eval_runtime": 4.1402,
+      "eval_samples_per_second": 67.87,
+      "eval_steps_per_second": 0.725,
+      "step": 747
+    },
+    {
+      "epoch": 230.77,
+      "eval_accuracy": 0.5729537366548043,
+      "eval_loss": 1.2600898742675781,
+      "eval_runtime": 3.9785,
+      "eval_samples_per_second": 70.63,
+      "eval_steps_per_second": 0.754,
+      "step": 750
+    },
+    {
+      "epoch": 232.0,
+      "eval_accuracy": 0.5729537366548043,
+      "eval_loss": 1.2625129222869873,
+      "eval_runtime": 4.1458,
+      "eval_samples_per_second": 67.779,
+      "eval_steps_per_second": 0.724,
+      "step": 754
+    },
+    {
+      "epoch": 232.92,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2635974884033203,
+      "eval_runtime": 4.5032,
+      "eval_samples_per_second": 62.401,
+      "eval_steps_per_second": 0.666,
+      "step": 757
+    },
+    {
+      "epoch": 233.85,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2629433870315552,
+      "eval_runtime": 4.1399,
+      "eval_samples_per_second": 67.876,
+      "eval_steps_per_second": 0.725,
+      "step": 760
+    },
+    {
+      "epoch": 234.77,
+      "eval_accuracy": 0.5765124555160143,
+      "eval_loss": 1.2600425481796265,
+      "eval_runtime": 4.2407,
+      "eval_samples_per_second": 66.263,
+      "eval_steps_per_second": 0.707,
+      "step": 763
+    },
+    {
+      "epoch": 236.0,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2558783292770386,
+      "eval_runtime": 4.1208,
+      "eval_samples_per_second": 68.19,
+      "eval_steps_per_second": 0.728,
+      "step": 767
+    },
+    {
+      "epoch": 236.92,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2534478902816772,
+      "eval_runtime": 3.8139,
+      "eval_samples_per_second": 73.678,
+      "eval_steps_per_second": 0.787,
+      "step": 770
+    },
+    {
+      "epoch": 237.85,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2513927221298218,
+      "eval_runtime": 4.6813,
+      "eval_samples_per_second": 60.026,
+      "eval_steps_per_second": 0.641,
+      "step": 773
+    },
+    {
+      "epoch": 238.77,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2508091926574707,
+      "eval_runtime": 4.2671,
+      "eval_samples_per_second": 65.852,
+      "eval_steps_per_second": 0.703,
+      "step": 776
+    },
+    {
+      "epoch": 240.0,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2487518787384033,
+      "eval_runtime": 3.7642,
+      "eval_samples_per_second": 74.651,
+      "eval_steps_per_second": 0.797,
+      "step": 780
+    },
+    {
+      "epoch": 240.92,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2483351230621338,
+      "eval_runtime": 4.8941,
+      "eval_samples_per_second": 57.416,
+      "eval_steps_per_second": 0.613,
+      "step": 783
+    },
+    {
+      "epoch": 241.85,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2500139474868774,
+      "eval_runtime": 4.274,
+      "eval_samples_per_second": 65.746,
+      "eval_steps_per_second": 0.702,
+      "step": 786
+    },
+    {
+      "epoch": 242.77,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2503968477249146,
+      "eval_runtime": 4.4982,
+      "eval_samples_per_second": 62.469,
+      "eval_steps_per_second": 0.667,
+      "step": 789
+    },
+    {
+      "epoch": 244.0,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2521419525146484,
+      "eval_runtime": 4.0413,
+      "eval_samples_per_second": 69.532,
+      "eval_steps_per_second": 0.742,
+      "step": 793
+    },
+    {
+      "epoch": 244.92,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2532862424850464,
+      "eval_runtime": 4.1262,
+      "eval_samples_per_second": 68.101,
+      "eval_steps_per_second": 0.727,
+      "step": 796
+    },
+    {
+      "epoch": 245.85,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.251287817955017,
+      "eval_runtime": 3.9321,
+      "eval_samples_per_second": 71.463,
+      "eval_steps_per_second": 0.763,
+      "step": 799
+    },
+    {
+      "epoch": 246.15,
+      "grad_norm": 63046.29296875,
+      "learning_rate": 3.7037037037037037e-06,
+      "loss": 0.5946,
+      "step": 800
+    },
+    {
+      "epoch": 246.77,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2513457536697388,
+      "eval_runtime": 4.1155,
+      "eval_samples_per_second": 68.279,
+      "eval_steps_per_second": 0.729,
+      "step": 802
+    },
+    {
+      "epoch": 248.0,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2507133483886719,
+      "eval_runtime": 4.3807,
+      "eval_samples_per_second": 64.145,
+      "eval_steps_per_second": 0.685,
+      "step": 806
+    },
+    {
+      "epoch": 248.92,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2491704225540161,
+      "eval_runtime": 4.0611,
+      "eval_samples_per_second": 69.193,
+      "eval_steps_per_second": 0.739,
+      "step": 809
+    },
+    {
+      "epoch": 249.85,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2499818801879883,
+      "eval_runtime": 3.9673,
+      "eval_samples_per_second": 70.828,
+      "eval_steps_per_second": 0.756,
+      "step": 812
+    },
+    {
+      "epoch": 250.77,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2505466938018799,
+      "eval_runtime": 4.5211,
+      "eval_samples_per_second": 62.153,
+      "eval_steps_per_second": 0.664,
+      "step": 815
+    },
+    {
+      "epoch": 252.0,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2519145011901855,
+      "eval_runtime": 5.2859,
+      "eval_samples_per_second": 53.16,
+      "eval_steps_per_second": 0.568,
+      "step": 819
+    },
+    {
+      "epoch": 252.92,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.253113865852356,
+      "eval_runtime": 4.0658,
+      "eval_samples_per_second": 69.113,
+      "eval_steps_per_second": 0.738,
+      "step": 822
+    },
+    {
+      "epoch": 253.85,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2538248300552368,
+      "eval_runtime": 4.1084,
+      "eval_samples_per_second": 68.396,
+      "eval_steps_per_second": 0.73,
+      "step": 825
+    },
+    {
+      "epoch": 254.77,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2532281875610352,
+      "eval_runtime": 4.0615,
+      "eval_samples_per_second": 69.186,
+      "eval_steps_per_second": 0.739,
+      "step": 828
+    },
+    {
+      "epoch": 256.0,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.2527676820755005,
+      "eval_runtime": 4.6892,
+      "eval_samples_per_second": 59.925,
+      "eval_steps_per_second": 0.64,
+      "step": 832
+    },
+    {
+      "epoch": 256.92,
+      "eval_accuracy": 0.5800711743772242,
+      "eval_loss": 1.252835988998413,
+      "eval_runtime": 3.7759,
+      "eval_samples_per_second": 74.42,
+      "eval_steps_per_second": 0.795,
+      "step": 835
+    },
+    {
+      "epoch": 257.85,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2521347999572754,
+      "eval_runtime": 3.5788,
+      "eval_samples_per_second": 78.519,
+      "eval_steps_per_second": 0.838,
+      "step": 838
+    },
+    {
+      "epoch": 258.77,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.252551555633545,
+      "eval_runtime": 3.9885,
+      "eval_samples_per_second": 70.452,
+      "eval_steps_per_second": 0.752,
+      "step": 841
+    },
+    {
+      "epoch": 260.0,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2527978420257568,
+      "eval_runtime": 3.7855,
+      "eval_samples_per_second": 74.231,
+      "eval_steps_per_second": 0.792,
+      "step": 845
+    },
+    {
+      "epoch": 260.92,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2529038190841675,
+      "eval_runtime": 3.8445,
+      "eval_samples_per_second": 73.091,
+      "eval_steps_per_second": 0.78,
+      "step": 848
+    },
+    {
+      "epoch": 261.85,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2528387308120728,
+      "eval_runtime": 3.7891,
+      "eval_samples_per_second": 74.16,
+      "eval_steps_per_second": 0.792,
+      "step": 851
+    },
+    {
+      "epoch": 262.77,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2516640424728394,
+      "eval_runtime": 4.2868,
+      "eval_samples_per_second": 65.55,
+      "eval_steps_per_second": 0.7,
+      "step": 854
+    },
+    {
+      "epoch": 264.0,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.251232385635376,
+      "eval_runtime": 3.7886,
+      "eval_samples_per_second": 74.169,
+      "eval_steps_per_second": 0.792,
+      "step": 858
+    },
+    {
+      "epoch": 264.92,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.251160979270935,
+      "eval_runtime": 4.4131,
+      "eval_samples_per_second": 63.674,
+      "eval_steps_per_second": 0.68,
+      "step": 861
+    },
+    {
+      "epoch": 265.85,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2503511905670166,
+      "eval_runtime": 3.8431,
+      "eval_samples_per_second": 73.118,
+      "eval_steps_per_second": 0.781,
+      "step": 864
+    },
+    {
+      "epoch": 266.77,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2499034404754639,
+      "eval_runtime": 3.7109,
+      "eval_samples_per_second": 75.723,
+      "eval_steps_per_second": 0.808,
+      "step": 867
+    },
+    {
+      "epoch": 268.0,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2496285438537598,
+      "eval_runtime": 4.0777,
+      "eval_samples_per_second": 68.912,
+      "eval_steps_per_second": 0.736,
+      "step": 871
+    },
+    {
+      "epoch": 268.92,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2497419118881226,
+      "eval_runtime": 4.4854,
+      "eval_samples_per_second": 62.648,
+      "eval_steps_per_second": 0.669,
+      "step": 874
+    },
+    {
+      "epoch": 269.85,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2500321865081787,
+      "eval_runtime": 5.2487,
+      "eval_samples_per_second": 53.537,
+      "eval_steps_per_second": 0.572,
+      "step": 877
+    },
+    {
+      "epoch": 270.77,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.250011682510376,
+      "eval_runtime": 4.3951,
+      "eval_samples_per_second": 63.935,
+      "eval_steps_per_second": 0.683,
+      "step": 880
+    },
+    {
+      "epoch": 272.0,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2498865127563477,
+      "eval_runtime": 4.1755,
+      "eval_samples_per_second": 67.297,
+      "eval_steps_per_second": 0.718,
+      "step": 884
+    },
+    {
+      "epoch": 272.92,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2500803470611572,
+      "eval_runtime": 4.6562,
+      "eval_samples_per_second": 60.349,
+      "eval_steps_per_second": 0.644,
+      "step": 887
+    },
+    {
+      "epoch": 273.85,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2503583431243896,
+      "eval_runtime": 4.5464,
+      "eval_samples_per_second": 61.807,
+      "eval_steps_per_second": 0.66,
+      "step": 890
+    },
+    {
+      "epoch": 274.77,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2506159543991089,
+      "eval_runtime": 3.9624,
+      "eval_samples_per_second": 70.917,
+      "eval_steps_per_second": 0.757,
+      "step": 893
+    },
+    {
+      "epoch": 276.0,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2505924701690674,
+      "eval_runtime": 4.0033,
+      "eval_samples_per_second": 70.192,
+      "eval_steps_per_second": 0.749,
+      "step": 897
+    },
+    {
+      "epoch": 276.92,
+      "grad_norm": 30700.47265625,
+      "learning_rate": 0.0,
+      "loss": 0.588,
+      "step": 900
+    },
+    {
+      "epoch": 276.92,
+      "eval_accuracy": 0.5836298932384342,
+      "eval_loss": 1.2505559921264648,
+      "eval_runtime": 4.7936,
+      "eval_samples_per_second": 58.62,
+      "eval_steps_per_second": 0.626,
+      "step": 900
+    },
+    {
+      "epoch": 276.92,
+      "step": 900,
+      "total_flos": 3.755576946691584e+18,
+      "train_loss": 0.8810926691691081,
+      "train_runtime": 3759.0782,
+      "train_samples_per_second": 123.541,
+      "train_steps_per_second": 0.239
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 900,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 300,
+  "save_steps": 500,
+  "total_flos": 3.755576946691584e+18,
+  "train_batch_size": 128,
+  "trial_name": null,
+  "trial_params": null
+}