{
  "best_metric": 0.5985733866691589,
  "best_model_checkpoint": "data/Llama-31-8B_task-2_60-samples_config-2_auto/checkpoint-25",
  "epoch": 16.0,
  "eval_steps": 500,
  "global_step": 46,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.3837566673755646,
      "learning_rate": 1e-05,
      "loss": 0.8962,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.3758746385574341,
      "learning_rate": 2e-05,
      "loss": 0.8574,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 0.8903820514678955,
      "eval_runtime": 13.347,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.38522425293922424,
      "learning_rate": 4e-05,
      "loss": 0.8746,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 0.8490483164787292,
      "eval_runtime": 13.3435,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 0.34568658471107483,
      "learning_rate": 6e-05,
      "loss": 0.8584,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.31414464116096497,
      "learning_rate": 8e-05,
      "loss": 0.7842,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 0.771674394607544,
      "eval_runtime": 13.3469,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.32403564453125,
      "learning_rate": 0.0001,
      "loss": 0.728,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 0.7139511108398438,
      "eval_runtime": 13.3421,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.3469023108482361,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.6997,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.31055331230163574,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.6405,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 0.6586517691612244,
      "eval_runtime": 13.3436,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.2623463273048401,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.5991,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.6365043520927429,
      "eval_runtime": 13.3397,
      "eval_samples_per_second": 0.9,
      "eval_steps_per_second": 0.9,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.26579177379608154,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.5578,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.26274964213371277,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.5371,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.6146900057792664,
      "eval_runtime": 13.3427,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.2567163407802582,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.5007,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.6017631888389587,
      "eval_runtime": 13.3431,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.2583542466163635,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.4783,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 0.5985733866691589,
      "eval_runtime": 13.346,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.26865774393081665,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.459,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.2833639979362488,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.4118,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 0.6050412058830261,
      "eval_runtime": 13.3451,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.290280818939209,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.4002,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 0.6277570128440857,
      "eval_runtime": 13.3488,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.3147587776184082,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.3665,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.31184399127960205,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.3234,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 0.6697214245796204,
      "eval_runtime": 13.3457,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.347219854593277,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.2816,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 0.7073009610176086,
      "eval_runtime": 13.3437,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.47505876421928406,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.2706,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.3857143223285675,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.2159,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 0.719702959060669,
      "eval_runtime": 13.3447,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 40
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.3344462513923645,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.1785,
      "step": 42
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 0.7951470017433167,
      "eval_runtime": 13.3459,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 43
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.3927725553512573,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.1504,
      "step": 44
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.42083150148391724,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.1209,
      "step": 46
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.9000644683837891,
      "eval_runtime": 13.3444,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.899,
      "step": 46
    },
    {
      "epoch": 16.0,
      "step": 46,
      "total_flos": 7.404745154730394e+16,
      "train_loss": 0.49191281426212063,
      "train_runtime": 2654.7246,
      "train_samples_per_second": 0.866,
      "train_steps_per_second": 0.038
    }
  ],
  "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": 7.404745154730394e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}