{
  "best_metric": 0.536961019039154,
  "best_model_checkpoint": "data/Mistral-7B_task-2_60-samples_config-2/checkpoint-20",
  "epoch": 13.91304347826087,
  "eval_steps": 500,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 3.3231360912323,
      "learning_rate": 1e-05,
      "loss": 1.023,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 2.4147260189056396,
      "learning_rate": 2e-05,
      "loss": 0.9782,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 0.9433401226997375,
      "eval_runtime": 16.1143,
      "eval_samples_per_second": 0.745,
      "eval_steps_per_second": 0.745,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 2.1237215995788574,
      "learning_rate": 4e-05,
      "loss": 0.9325,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 0.7483310699462891,
      "eval_runtime": 16.1136,
      "eval_samples_per_second": 0.745,
      "eval_steps_per_second": 0.745,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 1.2565068006515503,
      "learning_rate": 6e-05,
      "loss": 0.798,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.6507286429405212,
      "learning_rate": 8e-05,
      "loss": 0.657,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 0.6179646253585815,
      "eval_runtime": 16.314,
      "eval_samples_per_second": 0.736,
      "eval_steps_per_second": 0.736,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.5383930206298828,
      "learning_rate": 0.0001,
      "loss": 0.5682,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 0.594241738319397,
      "eval_runtime": 16.3164,
      "eval_samples_per_second": 0.735,
      "eval_steps_per_second": 0.735,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.5266416072845459,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.5826,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.4931633472442627,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.5483,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 0.5666075348854065,
      "eval_runtime": 16.3194,
      "eval_samples_per_second": 0.735,
      "eval_steps_per_second": 0.735,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.3791971504688263,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.4875,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.5440824627876282,
      "eval_runtime": 16.1338,
      "eval_samples_per_second": 0.744,
      "eval_steps_per_second": 0.744,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.3482791483402252,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.4524,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.3163185715675354,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.4347,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.536961019039154,
      "eval_runtime": 16.1301,
      "eval_samples_per_second": 0.744,
      "eval_steps_per_second": 0.744,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.3439599573612213,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.3995,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.5380921363830566,
      "eval_runtime": 16.1144,
      "eval_samples_per_second": 0.745,
      "eval_steps_per_second": 0.745,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.3523787558078766,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.3709,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 0.5430774688720703,
      "eval_runtime": 16.1301,
      "eval_samples_per_second": 0.744,
      "eval_steps_per_second": 0.744,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.4210048019886017,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.3236,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.4124290943145752,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.2801,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 0.5705265998840332,
      "eval_runtime": 16.2742,
      "eval_samples_per_second": 0.737,
      "eval_steps_per_second": 0.737,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.41128212213516235,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.2582,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 0.6300494074821472,
      "eval_runtime": 16.3217,
      "eval_samples_per_second": 0.735,
      "eval_steps_per_second": 0.735,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.6074411869049072,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.2046,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.5602520704269409,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.1682,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 0.6605353951454163,
      "eval_runtime": 16.305,
      "eval_samples_per_second": 0.736,
      "eval_steps_per_second": 0.736,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.6049512624740601,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.1192,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 0.7692939639091492,
      "eval_runtime": 16.1886,
      "eval_samples_per_second": 0.741,
      "eval_steps_per_second": 0.741,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.9413750171661377,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.0889,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.710867702960968,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0549,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 0.8641250729560852,
      "eval_runtime": 16.1382,
      "eval_samples_per_second": 0.744,
      "eval_steps_per_second": 0.744,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "step": 40,
      "total_flos": 7.711052122397082e+16,
      "train_loss": 0.4367667862446979,
      "train_runtime": 2829.7549,
      "train_samples_per_second": 0.813,
      "train_steps_per_second": 0.035
    }
  ],
  "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": 7.711052122397082e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}