{
  "best_metric": 0.6937975883483887,
  "best_model_checkpoint": "data/Gemma-2-2B_task-2_60-samples_config-2_auto/checkpoint-28",
  "epoch": 16.695652173913043,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.7196612358093262,
      "learning_rate": 1e-05,
      "loss": 1.0245,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.6809248924255371,
      "learning_rate": 2e-05,
      "loss": 0.9552,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 1.0745519399642944,
      "eval_runtime": 5.4065,
      "eval_samples_per_second": 2.22,
      "eval_steps_per_second": 2.22,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.6720972657203674,
      "learning_rate": 4e-05,
      "loss": 0.9555,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 1.032631278038025,
      "eval_runtime": 5.1616,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 0.6311763525009155,
      "learning_rate": 6e-05,
      "loss": 0.9764,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.5127931237220764,
      "learning_rate": 8e-05,
      "loss": 0.8712,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 0.9149246215820312,
      "eval_runtime": 5.1677,
      "eval_samples_per_second": 2.322,
      "eval_steps_per_second": 2.322,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.5144202709197998,
      "learning_rate": 0.0001,
      "loss": 0.7712,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 0.831948459148407,
      "eval_runtime": 5.1637,
      "eval_samples_per_second": 2.324,
      "eval_steps_per_second": 2.324,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.425078809261322,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.7448,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.42148834466934204,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.6958,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 0.7876434326171875,
      "eval_runtime": 5.1596,
      "eval_samples_per_second": 2.326,
      "eval_steps_per_second": 2.326,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.3883320987224579,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.6428,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.7528664469718933,
      "eval_runtime": 5.1626,
      "eval_samples_per_second": 2.324,
      "eval_steps_per_second": 2.324,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.3374294638633728,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.5915,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.3409869074821472,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.5843,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.7261620163917542,
      "eval_runtime": 5.1657,
      "eval_samples_per_second": 2.323,
      "eval_steps_per_second": 2.323,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.3142615854740143,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.561,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.7110547423362732,
      "eval_runtime": 5.1618,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.32146942615509033,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.5211,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 0.702172577381134,
      "eval_runtime": 5.1626,
      "eval_samples_per_second": 2.324,
      "eval_steps_per_second": 2.324,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.35450097918510437,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.5081,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.30655330419540405,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.456,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 0.6937975883483887,
      "eval_runtime": 5.1602,
      "eval_samples_per_second": 2.326,
      "eval_steps_per_second": 2.326,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.307801753282547,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.4502,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 0.6949651837348938,
      "eval_runtime": 5.1617,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.3407575488090515,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.4285,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.4042392373085022,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.3993,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 0.7011234164237976,
      "eval_runtime": 5.1622,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.3360240161418915,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.3589,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 0.7186114192008972,
      "eval_runtime": 5.1629,
      "eval_samples_per_second": 2.324,
      "eval_steps_per_second": 2.324,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.4053281545639038,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.3506,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.36633118987083435,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.3157,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 0.7445299029350281,
      "eval_runtime": 5.1647,
      "eval_samples_per_second": 2.323,
      "eval_steps_per_second": 2.323,
      "step": 40
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.35974082350730896,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.2717,
      "step": 42
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 0.7838297486305237,
      "eval_runtime": 5.1664,
      "eval_samples_per_second": 2.323,
      "eval_steps_per_second": 2.323,
      "step": 43
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.3751522898674011,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.2637,
      "step": 44
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.42833226919174194,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.2344,
      "step": 46
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.8412135243415833,
      "eval_runtime": 5.1574,
      "eval_samples_per_second": 2.327,
      "eval_steps_per_second": 2.327,
      "step": 46
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 0.3998011648654938,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.1863,
      "step": 48
    },
    {
      "epoch": 16.695652173913043,
      "eval_loss": 0.8865616917610168,
      "eval_runtime": 5.167,
      "eval_samples_per_second": 2.322,
      "eval_steps_per_second": 2.322,
      "step": 48
    },
    {
      "epoch": 16.695652173913043,
      "step": 48,
      "total_flos": 2.672993913484083e+16,
      "train_loss": 0.5509283027301232,
      "train_runtime": 1157.3749,
      "train_samples_per_second": 1.987,
      "train_steps_per_second": 0.086
    }
  ],
  "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": 2.672993913484083e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}