{
  "best_metric": 0.9476106762886047,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_60-samples_config-2_full_auto/checkpoint-37",
  "epoch": 19.82608695652174,
  "eval_steps": 500,
  "global_step": 57,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 1.285793662071228,
      "learning_rate": 1e-05,
      "loss": 2.3068,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 1.2761372327804565,
      "learning_rate": 2e-05,
      "loss": 2.3291,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 2.214773416519165,
      "eval_runtime": 1.0947,
      "eval_samples_per_second": 10.962,
      "eval_steps_per_second": 10.962,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 1.301371455192566,
      "learning_rate": 4e-05,
      "loss": 2.3149,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 2.120266914367676,
      "eval_runtime": 1.0594,
      "eval_samples_per_second": 11.328,
      "eval_steps_per_second": 11.328,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 1.28001070022583,
      "learning_rate": 6e-05,
      "loss": 2.272,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 1.0247910022735596,
      "learning_rate": 8e-05,
      "loss": 2.0527,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 1.7994542121887207,
      "eval_runtime": 1.0855,
      "eval_samples_per_second": 11.055,
      "eval_steps_per_second": 11.055,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.8930116891860962,
      "learning_rate": 0.0001,
      "loss": 1.8287,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 1.492956280708313,
      "eval_runtime": 1.0687,
      "eval_samples_per_second": 11.229,
      "eval_steps_per_second": 11.229,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.8637903928756714,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.5774,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.8775559067726135,
      "learning_rate": 9.951340343707852e-05,
      "loss": 1.4013,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.3095505237579346,
      "eval_runtime": 1.0623,
      "eval_samples_per_second": 11.296,
      "eval_steps_per_second": 11.296,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.9027936458587646,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.2792,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 1.145402193069458,
      "eval_runtime": 1.0607,
      "eval_samples_per_second": 11.314,
      "eval_steps_per_second": 11.314,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.8070787191390991,
      "learning_rate": 9.806308479691595e-05,
      "loss": 1.1154,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.5553158521652222,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.0352,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 1.0580171346664429,
      "eval_runtime": 1.0616,
      "eval_samples_per_second": 11.304,
      "eval_steps_per_second": 11.304,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.42665135860443115,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.9755,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.0166171789169312,
      "eval_runtime": 1.0673,
      "eval_samples_per_second": 11.243,
      "eval_steps_per_second": 11.243,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.40814900398254395,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.9229,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 0.9997214674949646,
      "eval_runtime": 1.0749,
      "eval_samples_per_second": 11.164,
      "eval_steps_per_second": 11.164,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.4242000877857208,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.9122,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.42280644178390503,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.8475,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 0.9786109328269958,
      "eval_runtime": 1.0544,
      "eval_samples_per_second": 11.381,
      "eval_steps_per_second": 11.381,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.3887423276901245,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.8219,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 0.9607260227203369,
      "eval_runtime": 1.0578,
      "eval_samples_per_second": 11.344,
      "eval_steps_per_second": 11.344,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.4444013237953186,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.8224,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.4059256911277771,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.7598,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 0.949688196182251,
      "eval_runtime": 1.054,
      "eval_samples_per_second": 11.386,
      "eval_steps_per_second": 11.386,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.4026019275188446,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.7358,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 0.9476106762886047,
      "eval_runtime": 1.0558,
      "eval_samples_per_second": 11.365,
      "eval_steps_per_second": 11.365,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.5041373372077942,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.71,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.4363657236099243,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.6627,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 0.9561586380004883,
      "eval_runtime": 1.0657,
      "eval_samples_per_second": 11.261,
      "eval_steps_per_second": 11.261,
      "step": 40
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.44975125789642334,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.6456,
      "step": 42
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 0.9710133671760559,
      "eval_runtime": 1.0594,
      "eval_samples_per_second": 11.327,
      "eval_steps_per_second": 11.327,
      "step": 43
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.4603601098060608,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.6009,
      "step": 44
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.48348623514175415,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.5933,
      "step": 46
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.9922751784324646,
      "eval_runtime": 1.0519,
      "eval_samples_per_second": 11.408,
      "eval_steps_per_second": 11.408,
      "step": 46
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 0.5055673122406006,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.5504,
      "step": 48
    },
    {
      "epoch": 16.695652173913043,
      "eval_loss": 1.0152024030685425,
      "eval_runtime": 1.0614,
      "eval_samples_per_second": 11.306,
      "eval_steps_per_second": 11.306,
      "step": 48
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 0.4959231913089752,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.5298,
      "step": 50
    },
    {
      "epoch": 17.73913043478261,
      "eval_loss": 1.054592251777649,
      "eval_runtime": 1.0666,
      "eval_samples_per_second": 11.251,
      "eval_steps_per_second": 11.251,
      "step": 51
    },
    {
      "epoch": 18.08695652173913,
      "grad_norm": 0.5696326494216919,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.4738,
      "step": 52
    },
    {
      "epoch": 18.782608695652176,
      "grad_norm": 0.5654062032699585,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.4504,
      "step": 54
    },
    {
      "epoch": 18.782608695652176,
      "eval_loss": 1.1058413982391357,
      "eval_runtime": 1.058,
      "eval_samples_per_second": 11.342,
      "eval_steps_per_second": 11.342,
      "step": 54
    },
    {
      "epoch": 19.47826086956522,
      "grad_norm": 0.5877916812896729,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.4508,
      "step": 56
    },
    {
      "epoch": 19.82608695652174,
      "eval_loss": 1.1509839296340942,
      "eval_runtime": 1.0593,
      "eval_samples_per_second": 11.328,
      "eval_steps_per_second": 11.328,
      "step": 57
    },
    {
      "epoch": 19.82608695652174,
      "step": 57,
      "total_flos": 3927682766077952.0,
      "train_loss": 1.0507982456893252,
      "train_runtime": 408.1771,
      "train_samples_per_second": 5.635,
      "train_steps_per_second": 0.245
    }
  ],
  "logging_steps": 2,
  "max_steps": 100,
  "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": 3927682766077952.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}