{
  "best_metric": 0.3567696511745453,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_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": 2.5350377559661865,
      "learning_rate": 1e-05,
      "loss": 2.3936,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 2.348182439804077,
      "learning_rate": 2e-05,
      "loss": 2.2034,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 2.2399237155914307,
      "eval_runtime": 11.3007,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 2.286853313446045,
      "learning_rate": 4e-05,
      "loss": 2.1509,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 1.8617448806762695,
      "eval_runtime": 11.3,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 2.149461030960083,
      "learning_rate": 6e-05,
      "loss": 2.0141,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 1.9644540548324585,
      "learning_rate": 8e-05,
      "loss": 1.5862,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 1.111824870109558,
      "eval_runtime": 11.2977,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 1.344768762588501,
      "learning_rate": 0.0001,
      "loss": 0.9806,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 0.5886362195014954,
      "eval_runtime": 11.2938,
      "eval_samples_per_second": 1.063,
      "eval_steps_per_second": 1.063,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.8945172429084778,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.5955,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.5430197715759277,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.4601,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 0.47657203674316406,
      "eval_runtime": 11.295,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 3.2166378498077393,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.4788,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.4127931296825409,
      "eval_runtime": 11.2942,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.6228726506233215,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.3295,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 4.100551128387451,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.3828,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.3871936798095703,
      "eval_runtime": 11.295,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.49670448899269104,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.2646,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.3888191878795624,
      "eval_runtime": 11.2961,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.5850313901901245,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.349,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 0.3619535267353058,
      "eval_runtime": 11.2955,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.330903559923172,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.3438,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.5838506817817688,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.299,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 0.3567696511745453,
      "eval_runtime": 11.296,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.28239765763282776,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.2883,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 0.35834944248199463,
      "eval_runtime": 11.2958,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.43306589126586914,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.2648,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.2673513889312744,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.2187,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 0.3658212125301361,
      "eval_runtime": 11.2946,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.2675131857395172,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.2439,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 0.36573949456214905,
      "eval_runtime": 11.2964,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.2865472733974457,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.2301,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.2461458444595337,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.1651,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 0.37130746245384216,
      "eval_runtime": 11.2918,
      "eval_samples_per_second": 1.063,
      "eval_steps_per_second": 1.063,
      "step": 40
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.33987730741500854,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.2365,
      "step": 42
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 0.38105103373527527,
      "eval_runtime": 11.297,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 43
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.3001103699207306,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.1509,
      "step": 44
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.2945997416973114,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.1654,
      "step": 46
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.38509896397590637,
      "eval_runtime": 11.2954,
      "eval_samples_per_second": 1.062,
      "eval_steps_per_second": 1.062,
      "step": 46
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 0.39486730098724365,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.161,
      "step": 48
    },
    {
      "epoch": 16.695652173913043,
      "eval_loss": 0.39996445178985596,
      "eval_runtime": 11.292,
      "eval_samples_per_second": 1.063,
      "eval_steps_per_second": 1.063,
      "step": 48
    },
    {
      "epoch": 16.695652173913043,
      "step": 48,
      "total_flos": 6.263550025452749e+16,
      "train_loss": 0.6125447253386179,
      "train_runtime": 2312.6306,
      "train_samples_per_second": 0.995,
      "train_steps_per_second": 0.043
    }
  ],
  "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": 6.263550025452749e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}