{
  "best_metric": 5.987339019775391,
  "best_model_checkpoint": "/tmp/model/checkpoint-28929",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 28929,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 3.3252678880055314e-05,
      "loss": 7.5851,
      "step": 1928
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.815831925026886e-05,
      "loss": 7.5065,
      "step": 3856
    },
    {
      "epoch": 0.6,
      "learning_rate": 4.445575357197727e-05,
      "loss": 7.4174,
      "step": 5784
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.075318789368567e-05,
      "loss": 7.2781,
      "step": 7712
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.705062221539407e-05,
      "loss": 7.1909,
      "step": 9640
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.01302842919923774,
      "eval_f1_macro": 0.0015898978634044435,
      "eval_f1_micro": 0.01302842919923774,
      "eval_f1_weighted": 0.0029651118829264957,
      "eval_loss": 6.873417854309082,
      "eval_precision_macro": 0.0031800239823964863,
      "eval_precision_micro": 0.01302842919923774,
      "eval_precision_weighted": 0.006035453178511265,
      "eval_recall_macro": 0.006325348961380463,
      "eval_recall_micro": 0.01302842919923774,
      "eval_recall_weighted": 0.01302842919923774,
      "eval_runtime": 286.2147,
      "eval_samples_per_second": 269.514,
      "eval_steps_per_second": 16.847,
      "step": 9643
    },
    {
      "epoch": 1.2,
      "learning_rate": 3.3349976954985404e-05,
      "loss": 6.9943,
      "step": 11568
    },
    {
      "epoch": 1.4,
      "learning_rate": 2.965125211245967e-05,
      "loss": 6.9127,
      "step": 13496
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.5948686434168075e-05,
      "loss": 6.8314,
      "step": 15424
    },
    {
      "epoch": 1.8,
      "learning_rate": 2.2248041173759414e-05,
      "loss": 6.764,
      "step": 17352
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.8549316331233677e-05,
      "loss": 6.7025,
      "step": 19280
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.03419800619660612,
      "eval_f1_macro": 0.008205818499317724,
      "eval_f1_micro": 0.03419800619660612,
      "eval_f1_weighted": 0.013979619158222838,
      "eval_loss": 6.236871719360352,
      "eval_precision_macro": 0.013553163045565874,
      "eval_precision_micro": 0.03419800619660612,
      "eval_precision_weighted": 0.021435662494438602,
      "eval_recall_macro": 0.018734909790546733,
      "eval_recall_micro": 0.03419800619660612,
      "eval_recall_weighted": 0.03419800619660612,
      "eval_runtime": 287.3858,
      "eval_samples_per_second": 268.416,
      "eval_steps_per_second": 16.779,
      "step": 19286
    },
    {
      "epoch": 2.2,
      "learning_rate": 1.4846750652942081e-05,
      "loss": 6.5556,
      "step": 21208
    },
    {
      "epoch": 2.4,
      "learning_rate": 1.1146105392533415e-05,
      "loss": 6.519,
      "step": 23136
    },
    {
      "epoch": 2.6,
      "learning_rate": 7.443539714241819e-06,
      "loss": 6.4822,
      "step": 25064
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.7428944538331538e-06,
      "loss": 6.4467,
      "step": 26992
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.224919342448917e-08,
      "loss": 6.4481,
      "step": 28920
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.049443212901385813,
      "eval_f1_macro": 0.014272821191130344,
      "eval_f1_micro": 0.049443212901385813,
      "eval_f1_weighted": 0.024064270485240995,
      "eval_loss": 5.987339019775391,
      "eval_precision_macro": 0.021826631000654732,
      "eval_precision_micro": 0.049443212901385813,
      "eval_precision_weighted": 0.03411589816873522,
      "eval_recall_macro": 0.02794951571811764,
      "eval_recall_micro": 0.049443212901385813,
      "eval_recall_weighted": 0.049443212901385813,
      "eval_runtime": 281.7111,
      "eval_samples_per_second": 273.823,
      "eval_steps_per_second": 17.117,
      "step": 28929
    }
  ],
  "logging_steps": 1928,
  "max_steps": 28929,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 5.796723355039476e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}