{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 3e-05,
      "loss": 2.9219,
      "step": 1
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.05291319857312723,
      "eval_loss": 2.6484375,
      "eval_runtime": 1.5576,
      "eval_samples_per_second": 18.618,
      "eval_steps_per_second": 1.284,
      "step": 1
    },
    {
      "epoch": 0.2,
      "learning_rate": 3e-05,
      "loss": 2.6938,
      "step": 2
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.05291319857312723,
      "eval_loss": 2.6484375,
      "eval_runtime": 1.275,
      "eval_samples_per_second": 22.746,
      "eval_steps_per_second": 1.569,
      "step": 2
    },
    {
      "epoch": 0.3,
      "learning_rate": 2.9970400926424075e-05,
      "loss": 2.6365,
      "step": 3
    },
    {
      "epoch": 0.3,
      "eval_accuracy": 0.05601796802748051,
      "eval_loss": 2.55078125,
      "eval_runtime": 1.1095,
      "eval_samples_per_second": 26.137,
      "eval_steps_per_second": 1.803,
      "step": 3
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.988172051971717e-05,
      "loss": 2.5088,
      "step": 4
    },
    {
      "epoch": 0.4,
      "eval_accuracy": 0.056282203725723345,
      "eval_loss": 2.533203125,
      "eval_runtime": 1.0069,
      "eval_samples_per_second": 28.8,
      "eval_steps_per_second": 1.986,
      "step": 4
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9734308760930333e-05,
      "loss": 2.7297,
      "step": 5
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.05667855727308759,
      "eval_loss": 2.517578125,
      "eval_runtime": 1.0127,
      "eval_samples_per_second": 28.637,
      "eval_steps_per_second": 1.975,
      "step": 5
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.9528747416929467e-05,
      "loss": 2.9702,
      "step": 6
    },
    {
      "epoch": 0.6,
      "eval_accuracy": 0.05720702866957326,
      "eval_loss": 2.494140625,
      "eval_runtime": 1.2159,
      "eval_samples_per_second": 23.851,
      "eval_steps_per_second": 1.645,
      "step": 6
    },
    {
      "epoch": 0.7,
      "learning_rate": 2.9265847744427305e-05,
      "loss": 2.729,
      "step": 7
    },
    {
      "epoch": 0.7,
      "eval_accuracy": 0.05681067512220901,
      "eval_loss": 2.48828125,
      "eval_runtime": 1.0262,
      "eval_samples_per_second": 28.259,
      "eval_steps_per_second": 1.949,
      "step": 7
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.894664728832377e-05,
      "loss": 2.6172,
      "step": 8
    },
    {
      "epoch": 0.8,
      "eval_accuracy": 0.05780155899061963,
      "eval_loss": 2.478515625,
      "eval_runtime": 1.3127,
      "eval_samples_per_second": 22.092,
      "eval_steps_per_second": 1.524,
      "step": 8
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.8572405786990293e-05,
      "loss": 2.6428,
      "step": 9
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.058065794688862464,
      "eval_loss": 2.458984375,
      "eval_runtime": 1.0971,
      "eval_samples_per_second": 26.432,
      "eval_steps_per_second": 1.823,
      "step": 9
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.8144600200657953e-05,
      "loss": 2.5681,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.05899061963271238,
      "eval_loss": 2.435546875,
      "eval_runtime": 1.3171,
      "eval_samples_per_second": 22.018,
      "eval_steps_per_second": 1.518,
      "step": 10
    },
    {
      "epoch": 1.1,
      "learning_rate": 2.7664918882530227e-05,
      "loss": 2.1885,
      "step": 11
    },
    {
      "epoch": 1.1,
      "eval_accuracy": 0.05866032500990884,
      "eval_loss": 2.423828125,
      "eval_runtime": 1.2181,
      "eval_samples_per_second": 23.807,
      "eval_steps_per_second": 1.642,
      "step": 11
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 1.981,
      "step": 12
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.05872638393446954,
      "eval_loss": 2.421875,
      "eval_runtime": 1.6039,
      "eval_samples_per_second": 18.081,
      "eval_steps_per_second": 1.247,
      "step": 12
    },
    {
      "epoch": 1.3,
      "learning_rate": 2.655769864163684e-05,
      "loss": 1.8673,
      "step": 13
    },
    {
      "epoch": 1.3,
      "eval_accuracy": 0.0591227374818338,
      "eval_loss": 2.41796875,
      "eval_runtime": 1.0123,
      "eval_samples_per_second": 28.649,
      "eval_steps_per_second": 1.976,
      "step": 13
    },
    {
      "epoch": 1.4,
      "learning_rate": 2.5934529411321174e-05,
      "loss": 1.7321,
      "step": 14
    },
    {
      "epoch": 1.4,
      "eval_accuracy": 0.05958514995375875,
      "eval_loss": 2.41796875,
      "eval_runtime": 1.0152,
      "eval_samples_per_second": 28.567,
      "eval_steps_per_second": 1.97,
      "step": 14
    },
    {
      "epoch": 1.5,
      "learning_rate": 2.5268206588930332e-05,
      "loss": 1.6355,
      "step": 15
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.060113621350244416,
      "eval_loss": 2.41796875,
      "eval_runtime": 1.4219,
      "eval_samples_per_second": 20.396,
      "eval_steps_per_second": 1.407,
      "step": 15
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.4561359846230346e-05,
      "loss": 1.7758,
      "step": 16
    },
    {
      "epoch": 1.6,
      "eval_accuracy": 0.06017968027480513,
      "eval_loss": 2.419921875,
      "eval_runtime": 1.618,
      "eval_samples_per_second": 17.923,
      "eval_steps_per_second": 1.236,
      "step": 16
    },
    {
      "epoch": 1.7,
      "learning_rate": 2.3816778784387097e-05,
      "loss": 2.0162,
      "step": 17
    },
    {
      "epoch": 1.7,
      "eval_accuracy": 0.06050997489760867,
      "eval_loss": 2.408203125,
      "eval_runtime": 1.1225,
      "eval_samples_per_second": 25.835,
      "eval_steps_per_second": 1.782,
      "step": 17
    },
    {
      "epoch": 1.8,
      "learning_rate": 2.303740192468495e-05,
      "loss": 1.8037,
      "step": 18
    },
    {
      "epoch": 1.8,
      "eval_accuracy": 0.06050997489760867,
      "eval_loss": 2.396484375,
      "eval_runtime": 0.912,
      "eval_samples_per_second": 31.8,
      "eval_steps_per_second": 2.193,
      "step": 18
    },
    {
      "epoch": 1.9,
      "learning_rate": 2.222630511152573e-05,
      "loss": 1.7204,
      "step": 19
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.0607742105958515,
      "eval_loss": 2.375,
      "eval_runtime": 1.2202,
      "eval_samples_per_second": 23.767,
      "eval_steps_per_second": 1.639,
      "step": 19
    },
    {
      "epoch": 2.0,
      "learning_rate": 2.138668937347609e-05,
      "loss": 1.7831,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.060906328444972915,
      "eval_loss": 2.357421875,
      "eval_runtime": 1.0082,
      "eval_samples_per_second": 28.765,
      "eval_steps_per_second": 1.984,
      "step": 20
    },
    {
      "epoch": 2.1,
      "learning_rate": 2.052186829027017e-05,
      "loss": 1.299,
      "step": 21
    },
    {
      "epoch": 2.1,
      "eval_accuracy": 0.06163297661514071,
      "eval_loss": 2.349609375,
      "eval_runtime": 1.2114,
      "eval_samples_per_second": 23.94,
      "eval_steps_per_second": 1.651,
      "step": 21
    },
    {
      "epoch": 2.2,
      "learning_rate": 1.963525491562421e-05,
      "loss": 1.4463,
      "step": 22
    },
    {
      "epoch": 2.2,
      "eval_accuracy": 0.06196327123794425,
      "eval_loss": 2.349609375,
      "eval_runtime": 1.5179,
      "eval_samples_per_second": 19.105,
      "eval_steps_per_second": 1.318,
      "step": 22
    },
    {
      "epoch": 2.3,
      "learning_rate": 1.8730348307472828e-05,
      "loss": 1.1733,
      "step": 23
    },
    {
      "epoch": 2.3,
      "eval_accuracy": 0.061699035539701415,
      "eval_loss": 2.365234375,
      "eval_runtime": 1.3077,
      "eval_samples_per_second": 22.176,
      "eval_steps_per_second": 1.529,
      "step": 23
    },
    {
      "epoch": 2.4,
      "learning_rate": 1.781071971878587e-05,
      "loss": 1.1142,
      "step": 24
    },
    {
      "epoch": 2.4,
      "eval_accuracy": 0.06255780155899061,
      "eval_loss": 2.388671875,
      "eval_runtime": 1.1042,
      "eval_samples_per_second": 26.264,
      "eval_steps_per_second": 1.811,
      "step": 24
    },
    {
      "epoch": 2.5,
      "learning_rate": 1.6879998503464565e-05,
      "loss": 1.3107,
      "step": 25
    },
    {
      "epoch": 2.5,
      "eval_accuracy": 0.06268991940811204,
      "eval_loss": 2.421875,
      "eval_runtime": 0.9167,
      "eval_samples_per_second": 31.634,
      "eval_steps_per_second": 2.182,
      "step": 25
    },
    {
      "epoch": 2.6,
      "learning_rate": 1.5941857792939702e-05,
      "loss": 1.011,
      "step": 26
    },
    {
      "epoch": 2.6,
      "eval_accuracy": 0.06216144801162637,
      "eval_loss": 2.455078125,
      "eval_runtime": 0.9207,
      "eval_samples_per_second": 31.497,
      "eval_steps_per_second": 2.172,
      "step": 26
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.5e-05,
      "loss": 1.3403,
      "step": 27
    },
    {
      "epoch": 2.7,
      "eval_accuracy": 0.061566917690579995,
      "eval_loss": 2.4765625,
      "eval_runtime": 1.5266,
      "eval_samples_per_second": 18.997,
      "eval_steps_per_second": 1.31,
      "step": 27
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.5e-05,
      "loss": 1.3108,
      "step": 28
    },
    {
      "epoch": 2.8,
      "eval_accuracy": 0.061566917690579995,
      "eval_loss": 2.4765625,
      "eval_runtime": 1.252,
      "eval_samples_per_second": 23.163,
      "eval_steps_per_second": 1.597,
      "step": 28
    },
    {
      "epoch": 2.9,
      "learning_rate": 1.40581422070603e-05,
      "loss": 1.0076,
      "step": 29
    },
    {
      "epoch": 2.9,
      "eval_accuracy": 0.06189721231338354,
      "eval_loss": 2.4609375,
      "eval_runtime": 0.9112,
      "eval_samples_per_second": 31.825,
      "eval_steps_per_second": 2.195,
      "step": 29
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.3120001496535434e-05,
      "loss": 0.8656,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.062359624785308494,
      "eval_loss": 2.451171875,
      "eval_runtime": 1.5156,
      "eval_samples_per_second": 19.134,
      "eval_steps_per_second": 1.32,
      "step": 30
    },
    {
      "epoch": 3.1,
      "learning_rate": 1.2189280281214128e-05,
      "loss": 0.6635,
      "step": 31
    },
    {
      "epoch": 3.1,
      "eval_accuracy": 0.06282203725723345,
      "eval_loss": 2.451171875,
      "eval_runtime": 1.313,
      "eval_samples_per_second": 22.087,
      "eval_steps_per_second": 1.523,
      "step": 31
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.1269651692527181e-05,
      "loss": 0.9996,
      "step": 32
    },
    {
      "epoch": 3.2,
      "eval_accuracy": 0.06348262650284053,
      "eval_loss": 2.443359375,
      "eval_runtime": 1.01,
      "eval_samples_per_second": 28.711,
      "eval_steps_per_second": 1.98,
      "step": 32
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.036474508437579e-05,
      "loss": 0.9029,
      "step": 33
    },
    {
      "epoch": 3.3,
      "eval_accuracy": 0.06368080327652266,
      "eval_loss": 2.447265625,
      "eval_runtime": 1.4214,
      "eval_samples_per_second": 20.402,
      "eval_steps_per_second": 1.407,
      "step": 33
    },
    {
      "epoch": 3.4,
      "learning_rate": 9.478131709729831e-06,
      "loss": 0.8329,
      "step": 34
    },
    {
      "epoch": 3.4,
      "eval_accuracy": 0.06374686220108336,
      "eval_loss": 2.455078125,
      "eval_runtime": 0.9136,
      "eval_samples_per_second": 31.742,
      "eval_steps_per_second": 2.189,
      "step": 34
    },
    {
      "epoch": 3.5,
      "learning_rate": 8.61331062652391e-06,
      "loss": 0.8012,
      "step": 35
    },
    {
      "epoch": 3.5,
      "eval_accuracy": 0.06387898005020479,
      "eval_loss": 2.46484375,
      "eval_runtime": 1.6062,
      "eval_samples_per_second": 18.055,
      "eval_steps_per_second": 1.245,
      "step": 35
    },
    {
      "epoch": 3.6,
      "learning_rate": 7.773694888474268e-06,
      "loss": 0.5814,
      "step": 36
    },
    {
      "epoch": 3.6,
      "eval_accuracy": 0.0640110978993262,
      "eval_loss": 2.490234375,
      "eval_runtime": 1.6209,
      "eval_samples_per_second": 17.891,
      "eval_steps_per_second": 1.234,
      "step": 36
    },
    {
      "epoch": 3.7,
      "learning_rate": 6.962598075315047e-06,
      "loss": 1.0688,
      "step": 37
    },
    {
      "epoch": 3.7,
      "eval_accuracy": 0.06381292112564407,
      "eval_loss": 2.509765625,
      "eval_runtime": 1.2235,
      "eval_samples_per_second": 23.703,
      "eval_steps_per_second": 1.635,
      "step": 37
    },
    {
      "epoch": 3.8,
      "learning_rate": 6.1832212156129045e-06,
      "loss": 0.8688,
      "step": 38
    },
    {
      "epoch": 3.8,
      "eval_accuracy": 0.06348262650284053,
      "eval_loss": 2.517578125,
      "eval_runtime": 1.5088,
      "eval_samples_per_second": 19.221,
      "eval_steps_per_second": 1.326,
      "step": 38
    },
    {
      "epoch": 3.9,
      "learning_rate": 5.438640153769654e-06,
      "loss": 0.7341,
      "step": 39
    },
    {
      "epoch": 3.9,
      "eval_accuracy": 0.06381292112564407,
      "eval_loss": 2.51953125,
      "eval_runtime": 1.0209,
      "eval_samples_per_second": 28.406,
      "eval_steps_per_second": 1.959,
      "step": 39
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.731793411069669e-06,
      "loss": 0.7102,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.0640110978993262,
      "eval_loss": 2.51953125,
      "eval_runtime": 1.3112,
      "eval_samples_per_second": 22.116,
      "eval_steps_per_second": 1.525,
      "step": 40
    },
    {
      "epoch": 4.1,
      "learning_rate": 4.06547058867883e-06,
      "loss": 0.7079,
      "step": 41
    },
    {
      "epoch": 4.1,
      "eval_accuracy": 0.06414321574844761,
      "eval_loss": 2.51953125,
      "eval_runtime": 1.0046,
      "eval_samples_per_second": 28.868,
      "eval_steps_per_second": 1.991,
      "step": 41
    },
    {
      "epoch": 4.2,
      "learning_rate": 3.442301358363163e-06,
      "loss": 0.7656,
      "step": 42
    },
    {
      "epoch": 4.2,
      "eval_accuracy": 0.06427533359756903,
      "eval_loss": 2.51953125,
      "eval_runtime": 1.3174,
      "eval_samples_per_second": 22.012,
      "eval_steps_per_second": 1.518,
      "step": 42
    },
    {
      "epoch": 4.3,
      "learning_rate": 2.86474508437579e-06,
      "loss": 0.6377,
      "step": 43
    },
    {
      "epoch": 4.3,
      "eval_accuracy": 0.06447351037125115,
      "eval_loss": 2.52734375,
      "eval_runtime": 1.1043,
      "eval_samples_per_second": 26.261,
      "eval_steps_per_second": 1.811,
      "step": 43
    },
    {
      "epoch": 4.4,
      "learning_rate": 2.335081117469777e-06,
      "loss": 0.5898,
      "step": 44
    },
    {
      "epoch": 4.4,
      "eval_accuracy": 0.06414321574844761,
      "eval_loss": 2.53515625,
      "eval_runtime": 1.0226,
      "eval_samples_per_second": 28.359,
      "eval_steps_per_second": 1.956,
      "step": 44
    },
    {
      "epoch": 4.5,
      "learning_rate": 1.8553997993420495e-06,
      "loss": 0.5958,
      "step": 45
    },
    {
      "epoch": 4.5,
      "eval_accuracy": 0.06407715682388691,
      "eval_loss": 2.54296875,
      "eval_runtime": 0.91,
      "eval_samples_per_second": 31.867,
      "eval_steps_per_second": 2.198,
      "step": 45
    },
    {
      "epoch": 4.6,
      "learning_rate": 1.4275942130097097e-06,
      "loss": 0.7048,
      "step": 46
    },
    {
      "epoch": 4.6,
      "eval_accuracy": 0.0640110978993262,
      "eval_loss": 2.548828125,
      "eval_runtime": 1.2089,
      "eval_samples_per_second": 23.989,
      "eval_steps_per_second": 1.654,
      "step": 46
    },
    {
      "epoch": 4.7,
      "learning_rate": 1.0533527116762298e-06,
      "loss": 0.5435,
      "step": 47
    },
    {
      "epoch": 4.7,
      "eval_accuracy": 0.06414321574844761,
      "eval_loss": 2.552734375,
      "eval_runtime": 1.3311,
      "eval_samples_per_second": 21.786,
      "eval_steps_per_second": 1.502,
      "step": 47
    },
    {
      "epoch": 4.8,
      "learning_rate": 7.341522555726971e-07,
      "loss": 0.4769,
      "step": 48
    },
    {
      "epoch": 4.8,
      "eval_accuracy": 0.0640110978993262,
      "eval_loss": 2.552734375,
      "eval_runtime": 1.0089,
      "eval_samples_per_second": 28.743,
      "eval_steps_per_second": 1.982,
      "step": 48
    },
    {
      "epoch": 4.9,
      "learning_rate": 4.7125258307053385e-07,
      "loss": 0.6583,
      "step": 49
    },
    {
      "epoch": 4.9,
      "eval_accuracy": 0.06420927467300833,
      "eval_loss": 2.5546875,
      "eval_runtime": 1.0099,
      "eval_samples_per_second": 28.715,
      "eval_steps_per_second": 1.98,
      "step": 49
    },
    {
      "epoch": 5.0,
      "learning_rate": 2.6569123906967083e-07,
      "loss": 0.7168,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.06414321574844761,
      "eval_loss": 2.5546875,
      "eval_runtime": 1.5134,
      "eval_samples_per_second": 19.163,
      "eval_steps_per_second": 1.322,
      "step": 50
    },
    {
      "epoch": 5.0,
      "step": 50,
      "total_flos": 2477483753472.0,
      "train_loss": 1.4392181396484376,
      "train_runtime": 183.8573,
      "train_samples_per_second": 4.242,
      "train_steps_per_second": 0.272
    }
  ],
  "max_steps": 50,
  "num_train_epochs": 5,
  "total_flos": 2477483753472.0,
  "trial_name": null,
  "trial_params": null
}