{
  "best_metric": 0.9620824456214905,
  "best_model_checkpoint": "data/Gemma-2-2B_task-3_60-samples_config-1_full/checkpoint-51",
  "epoch": 16.0,
  "eval_steps": 500,
  "global_step": 92,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 0.42138275504112244,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.389,
      "step": 1
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.4729943871498108,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.4165,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.4365766942501068,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.3556,
      "step": 4
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 1.3616156578063965,
      "eval_runtime": 4.3266,
      "eval_samples_per_second": 2.774,
      "eval_steps_per_second": 2.774,
      "step": 5
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 0.4426564574241638,
      "learning_rate": 2.4e-05,
      "loss": 1.3728,
      "step": 6
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.4195796847343445,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.3641,
      "step": 8
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 0.327674925327301,
      "learning_rate": 4e-05,
      "loss": 1.3617,
      "step": 10
    },
    {
      "epoch": 1.9130434782608696,
      "eval_loss": 1.283636212348938,
      "eval_runtime": 4.152,
      "eval_samples_per_second": 2.89,
      "eval_steps_per_second": 2.89,
      "step": 11
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 0.366742342710495,
      "learning_rate": 4.8e-05,
      "loss": 1.3215,
      "step": 12
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 0.25038954615592957,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.2099,
      "step": 14
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.2523439824581146,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.257,
      "step": 16
    },
    {
      "epoch": 2.9565217391304346,
      "eval_loss": 1.181667447090149,
      "eval_runtime": 4.1479,
      "eval_samples_per_second": 2.893,
      "eval_steps_per_second": 2.893,
      "step": 17
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 0.25941434502601624,
      "learning_rate": 7.2e-05,
      "loss": 1.1446,
      "step": 18
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.3189179599285126,
      "learning_rate": 8e-05,
      "loss": 1.1484,
      "step": 20
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 0.2823261618614197,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.1091,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.0975886583328247,
      "eval_runtime": 4.1853,
      "eval_samples_per_second": 2.867,
      "eval_steps_per_second": 2.867,
      "step": 23
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.2734120488166809,
      "learning_rate": 9.6e-05,
      "loss": 1.0954,
      "step": 24
    },
    {
      "epoch": 4.521739130434782,
      "grad_norm": 0.3026849925518036,
      "learning_rate": 9.999512620046522e-05,
      "loss": 1.0489,
      "step": 26
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.2881958484649658,
      "learning_rate": 9.995614150494293e-05,
      "loss": 0.9986,
      "step": 28
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.030081033706665,
      "eval_runtime": 4.1926,
      "eval_samples_per_second": 2.862,
      "eval_steps_per_second": 2.862,
      "step": 28
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 0.2395625114440918,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.9491,
      "step": 30
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.21183127164840698,
      "learning_rate": 9.976136999909156e-05,
      "loss": 0.9619,
      "step": 32
    },
    {
      "epoch": 5.913043478260869,
      "grad_norm": 0.19583259522914886,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.9277,
      "step": 34
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.9902016520500183,
      "eval_runtime": 4.1944,
      "eval_samples_per_second": 2.861,
      "eval_steps_per_second": 2.861,
      "step": 34
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.19764377176761627,
      "learning_rate": 9.941141907232765e-05,
      "loss": 0.908,
      "step": 36
    },
    {
      "epoch": 6.608695652173913,
      "grad_norm": 0.23089689016342163,
      "learning_rate": 9.917857354066931e-05,
      "loss": 0.9093,
      "step": 38
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.1905813366174698,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.8595,
      "step": 40
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.9712240695953369,
      "eval_runtime": 4.1882,
      "eval_samples_per_second": 2.865,
      "eval_steps_per_second": 2.865,
      "step": 40
    },
    {
      "epoch": 7.304347826086957,
      "grad_norm": 0.7232856154441833,
      "learning_rate": 9.859805002892732e-05,
      "loss": 0.9108,
      "step": 42
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.2041950672864914,
      "learning_rate": 9.825082472361557e-05,
      "loss": 0.8481,
      "step": 44
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.21693262457847595,
      "learning_rate": 9.786597487660337e-05,
      "loss": 0.8452,
      "step": 46
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.9633083343505859,
      "eval_runtime": 4.1841,
      "eval_samples_per_second": 2.868,
      "eval_steps_per_second": 2.868,
      "step": 46
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.205765500664711,
      "learning_rate": 9.744380058222483e-05,
      "loss": 0.8451,
      "step": 48
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 0.22007052600383759,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.8318,
      "step": 50
    },
    {
      "epoch": 8.869565217391305,
      "eval_loss": 0.9620824456214905,
      "eval_runtime": 4.1935,
      "eval_samples_per_second": 2.862,
      "eval_steps_per_second": 2.862,
      "step": 51
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.22718147933483124,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.8169,
      "step": 52
    },
    {
      "epoch": 9.391304347826088,
      "grad_norm": 0.22475072741508484,
      "learning_rate": 9.595676696276172e-05,
      "loss": 0.7724,
      "step": 54
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.22813497483730316,
      "learning_rate": 9.538887392664544e-05,
      "loss": 0.8165,
      "step": 56
    },
    {
      "epoch": 9.91304347826087,
      "eval_loss": 0.9639017581939697,
      "eval_runtime": 4.1904,
      "eval_samples_per_second": 2.864,
      "eval_steps_per_second": 2.864,
      "step": 57
    },
    {
      "epoch": 10.08695652173913,
      "grad_norm": 0.23416085541248322,
      "learning_rate": 9.478558801197065e-05,
      "loss": 0.8024,
      "step": 58
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.25057733058929443,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.7687,
      "step": 60
    },
    {
      "epoch": 10.782608695652174,
      "grad_norm": 0.2932647168636322,
      "learning_rate": 9.347474647526095e-05,
      "loss": 0.7226,
      "step": 62
    },
    {
      "epoch": 10.956521739130435,
      "eval_loss": 0.9804084300994873,
      "eval_runtime": 4.1914,
      "eval_samples_per_second": 2.863,
      "eval_steps_per_second": 2.863,
      "step": 63
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.33015286922454834,
      "learning_rate": 9.276821300802534e-05,
      "loss": 0.7163,
      "step": 64
    },
    {
      "epoch": 11.478260869565217,
      "grad_norm": 0.3211749494075775,
      "learning_rate": 9.202833017478422e-05,
      "loss": 0.7702,
      "step": 66
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.35134953260421753,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.6818,
      "step": 68
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.0012708902359009,
      "eval_runtime": 4.1869,
      "eval_samples_per_second": 2.866,
      "eval_steps_per_second": 2.866,
      "step": 69
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 0.36197611689567566,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.6894,
      "step": 70
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.38006392121315,
      "learning_rate": 8.961448216775954e-05,
      "loss": 0.6214,
      "step": 72
    },
    {
      "epoch": 12.869565217391305,
      "grad_norm": 0.460814893245697,
      "learning_rate": 8.874722443520899e-05,
      "loss": 0.6785,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 1.024221658706665,
      "eval_runtime": 4.192,
      "eval_samples_per_second": 2.863,
      "eval_steps_per_second": 2.863,
      "step": 74
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.6328468918800354,
      "learning_rate": 8.784975278258783e-05,
      "loss": 0.6288,
      "step": 76
    },
    {
      "epoch": 13.565217391304348,
      "grad_norm": 0.5195140242576599,
      "learning_rate": 8.692276703129421e-05,
      "loss": 0.6211,
      "step": 78
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.5798870325088501,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.5736,
      "step": 80
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 1.0798345804214478,
      "eval_runtime": 4.187,
      "eval_samples_per_second": 2.866,
      "eval_steps_per_second": 2.866,
      "step": 80
    },
    {
      "epoch": 14.26086956521739,
      "grad_norm": 0.649289608001709,
      "learning_rate": 8.498316702566828e-05,
      "loss": 0.5203,
      "step": 82
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.7385919094085693,
      "learning_rate": 8.397206521307584e-05,
      "loss": 0.521,
      "step": 84
    },
    {
      "epoch": 14.956521739130435,
      "grad_norm": 0.7277963757514954,
      "learning_rate": 8.293447300593402e-05,
      "loss": 0.5264,
      "step": 86
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 1.1451330184936523,
      "eval_runtime": 4.1463,
      "eval_samples_per_second": 2.894,
      "eval_steps_per_second": 2.894,
      "step": 86
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.7127543687820435,
      "learning_rate": 8.18711994874345e-05,
      "loss": 0.4856,
      "step": 88
    },
    {
      "epoch": 15.652173913043478,
      "grad_norm": 0.9132648706436157,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.3871,
      "step": 90
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.8008753061294556,
      "learning_rate": 7.967094433018508e-05,
      "loss": 0.5243,
      "step": 92
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.2084286212921143,
      "eval_runtime": 4.1468,
      "eval_samples_per_second": 2.894,
      "eval_steps_per_second": 2.894,
      "step": 92
    },
    {
      "epoch": 16.0,
      "step": 92,
      "total_flos": 2.0083778575663104e+16,
      "train_loss": 0.8833115586768026,
      "train_runtime": 871.0167,
      "train_samples_per_second": 2.641,
      "train_steps_per_second": 0.287
    }
  ],
  "logging_steps": 2,
  "max_steps": 250,
  "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": 2.0083778575663104e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}