{
  "best_metric": 1.2883715629577637,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_180-samples_config-2/checkpoint-51",
  "epoch": 12.941176470588236,
  "eval_steps": 500,
  "global_step": 110,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 2.12469220161438,
      "learning_rate": 2.5e-06,
      "loss": 2.2628,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 2.373323917388916,
      "learning_rate": 5e-06,
      "loss": 2.3042,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 2.6006672382354736,
      "learning_rate": 1e-05,
      "loss": 2.3721,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 2.6902971267700195,
      "learning_rate": 1.5e-05,
      "loss": 2.4244,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 2.508831262588501,
      "learning_rate": 2e-05,
      "loss": 2.3482,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 2.2692689895629883,
      "eval_runtime": 3.3884,
      "eval_samples_per_second": 10.625,
      "eval_steps_per_second": 10.625,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 2.0970139503479004,
      "learning_rate": 2.5e-05,
      "loss": 2.2452,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 1.9512650966644287,
      "learning_rate": 3e-05,
      "loss": 2.1602,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 2.0027029514312744,
      "learning_rate": 3.5e-05,
      "loss": 2.0043,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 1.7758914232254028,
      "learning_rate": 4e-05,
      "loss": 1.7809,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.6932330131530762,
      "eval_runtime": 3.3144,
      "eval_samples_per_second": 10.862,
      "eval_steps_per_second": 10.862,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 1.2240898609161377,
      "learning_rate": 4.5e-05,
      "loss": 1.6164,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 1.2644752264022827,
      "learning_rate": 5e-05,
      "loss": 1.5868,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.9468474984169006,
      "learning_rate": 5.500000000000001e-05,
      "loss": 1.5945,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.8889049291610718,
      "learning_rate": 6e-05,
      "loss": 1.553,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 1.5041224956512451,
      "eval_runtime": 3.3098,
      "eval_samples_per_second": 10.877,
      "eval_steps_per_second": 10.877,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.966974675655365,
      "learning_rate": 6.500000000000001e-05,
      "loss": 1.4543,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.7903351783752441,
      "learning_rate": 7e-05,
      "loss": 1.3984,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.7399197220802307,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.3838,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.7915351390838623,
      "learning_rate": 8e-05,
      "loss": 1.2771,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8471887111663818,
      "learning_rate": 8.5e-05,
      "loss": 1.2823,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.3677548170089722,
      "eval_runtime": 3.405,
      "eval_samples_per_second": 10.573,
      "eval_steps_per_second": 10.573,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.7967553734779358,
      "learning_rate": 9e-05,
      "loss": 1.2545,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.8264508843421936,
      "learning_rate": 9.5e-05,
      "loss": 1.1629,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.9275144338607788,
      "learning_rate": 0.0001,
      "loss": 1.1813,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.9521178603172302,
      "learning_rate": 9.999238475781957e-05,
      "loss": 1.1298,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 1.3192731142044067,
      "eval_runtime": 3.1829,
      "eval_samples_per_second": 11.31,
      "eval_steps_per_second": 11.31,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.8590582609176636,
      "learning_rate": 9.99695413509548e-05,
      "loss": 1.0744,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.8991239666938782,
      "learning_rate": 9.99314767377287e-05,
      "loss": 1.0672,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 1.0707749128341675,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.9493,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 1.1669797897338867,
      "learning_rate": 9.980973490458728e-05,
      "loss": 1.027,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.2883715629577637,
      "eval_runtime": 3.1944,
      "eval_samples_per_second": 11.27,
      "eval_steps_per_second": 11.27,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.9919552803039551,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.9344,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.9965695738792419,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.8502,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 1.1732594966888428,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.8699,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 1.213049292564392,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.8219,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 1.3589649200439453,
      "eval_runtime": 3.135,
      "eval_samples_per_second": 11.483,
      "eval_steps_per_second": 11.483,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 1.3312314748764038,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.7914,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 1.2881100177764893,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.713,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 1.4042942523956299,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.6519,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 1.5494970083236694,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.5786,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.7536710500717163,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.6244,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.5295920372009277,
      "eval_runtime": 3.1858,
      "eval_samples_per_second": 11.3,
      "eval_steps_per_second": 11.3,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 1.5150338411331177,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.4796,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 1.9343677759170532,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.5302,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 2.203670024871826,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.4138,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 2.373121738433838,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.398,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 1.7941030263900757,
      "eval_runtime": 3.121,
      "eval_samples_per_second": 11.535,
      "eval_steps_per_second": 11.535,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 1.649604320526123,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.3333,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 2.0595147609710693,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.2576,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 2.2983481884002686,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.2559,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 2.6052489280700684,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.2704,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.2475032806396484,
      "eval_runtime": 3.119,
      "eval_samples_per_second": 11.542,
      "eval_steps_per_second": 11.542,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 2.4219186305999756,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.2105,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 1.5558136701583862,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.1226,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 2.0058951377868652,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.1645,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 1.9732674360275269,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.1592,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 2.5631463527679443,
      "eval_runtime": 3.2,
      "eval_samples_per_second": 11.25,
      "eval_steps_per_second": 11.25,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 2.4981675148010254,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.1372,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 1.9884262084960938,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0687,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 1.781720519065857,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.0861,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 2.489872932434082,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0753,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 3.094914674758911,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.0797,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 2.784686326980591,
      "eval_runtime": 3.1247,
      "eval_samples_per_second": 11.521,
      "eval_steps_per_second": 11.521,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 1.77165949344635,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0486,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 1.576274037361145,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0495,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 1.8895446062088013,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0434,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 1.6305781602859497,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.0459,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 2.944411277770996,
      "eval_runtime": 3.1195,
      "eval_samples_per_second": 11.54,
      "eval_steps_per_second": 11.54,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "step": 110,
      "total_flos": 7344862462476288.0,
      "train_loss": 0.9215991622006351,
      "train_runtime": 758.0034,
      "train_samples_per_second": 8.971,
      "train_steps_per_second": 0.528
    }
  ],
  "logging_steps": 2,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7344862462476288.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}