{
  "best_metric": 1.2879904508590698,
  "best_model_checkpoint": "data/Llama-31-8B_task-1_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": 1.842164158821106,
      "learning_rate": 1e-05,
      "loss": 2.0791,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 1.7369345426559448,
      "learning_rate": 2e-05,
      "loss": 2.1668,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 2.061022996902466,
      "eval_runtime": 4.8455,
      "eval_samples_per_second": 2.477,
      "eval_steps_per_second": 2.477,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 1.8460229635238647,
      "learning_rate": 4e-05,
      "loss": 2.1104,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 1.8653119802474976,
      "eval_runtime": 4.844,
      "eval_samples_per_second": 2.477,
      "eval_steps_per_second": 2.477,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 1.571470856666565,
      "learning_rate": 6e-05,
      "loss": 1.8563,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 1.0414882898330688,
      "learning_rate": 8e-05,
      "loss": 1.6856,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 1.5622197389602661,
      "eval_runtime": 4.8463,
      "eval_samples_per_second": 2.476,
      "eval_steps_per_second": 2.476,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.7915772199630737,
      "learning_rate": 0.0001,
      "loss": 1.4548,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 1.4690589904785156,
      "eval_runtime": 4.8526,
      "eval_samples_per_second": 2.473,
      "eval_steps_per_second": 2.473,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.7557133436203003,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.4815,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.7220547795295715,
      "learning_rate": 9.951340343707852e-05,
      "loss": 1.3644,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.3790850639343262,
      "eval_runtime": 4.8503,
      "eval_samples_per_second": 2.474,
      "eval_steps_per_second": 2.474,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.578528642654419,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.2755,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 1.3223717212677002,
      "eval_runtime": 4.8439,
      "eval_samples_per_second": 2.477,
      "eval_steps_per_second": 2.477,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.5845955014228821,
      "learning_rate": 9.806308479691595e-05,
      "loss": 1.1074,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.5618941187858582,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.1243,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 1.2879904508590698,
      "eval_runtime": 4.8446,
      "eval_samples_per_second": 2.477,
      "eval_steps_per_second": 2.477,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.6283526420593262,
      "learning_rate": 9.567727288213005e-05,
      "loss": 1.0234,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.3072248697280884,
      "eval_runtime": 4.842,
      "eval_samples_per_second": 2.478,
      "eval_steps_per_second": 2.478,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.7448239326477051,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.933,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 1.3175863027572632,
      "eval_runtime": 4.8516,
      "eval_samples_per_second": 2.473,
      "eval_steps_per_second": 2.473,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.8109312057495117,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.8709,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.8030046820640564,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.7437,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 1.4185105562210083,
      "eval_runtime": 4.8338,
      "eval_samples_per_second": 2.483,
      "eval_steps_per_second": 2.483,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 1.1518187522888184,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.6589,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 1.4575523138046265,
      "eval_runtime": 4.8224,
      "eval_samples_per_second": 2.488,
      "eval_steps_per_second": 2.488,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 1.1305478811264038,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.6139,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.9890539050102234,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.4603,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 1.5914921760559082,
      "eval_runtime": 4.8278,
      "eval_samples_per_second": 2.486,
      "eval_steps_per_second": 2.486,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 1.2683886289596558,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.3502,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 1.7685762643814087,
      "eval_runtime": 4.8299,
      "eval_samples_per_second": 2.485,
      "eval_steps_per_second": 2.485,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 1.8165357112884521,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.3095,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 1.0062901973724365,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.1758,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 1.9850009679794312,
      "eval_runtime": 4.8261,
      "eval_samples_per_second": 2.486,
      "eval_steps_per_second": 2.486,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "step": 40,
      "total_flos": 9639644377382912.0,
      "train_loss": 1.0869066467974335,
      "train_runtime": 755.7673,
      "train_samples_per_second": 3.043,
      "train_steps_per_second": 0.132
    }
  ],
  "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": 9639644377382912.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}