{
  "best_metric": 0.5653966665267944,
  "best_model_checkpoint": "/mnt/beegfs/farid/mlora/outputs/xnli/aya-101/sw/rank4_lr5e-5/checkpoint-5500",
  "epoch": 0.24445893089960888,
  "eval_steps": 500,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.020371577574967405,
      "grad_norm": 2.0201494693756104,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.1486,
      "step": 500
    },
    {
      "epoch": 0.020371577574967405,
      "eval_accuracy": 0.37309236947791163,
      "eval_f1": 0.3322283925001309,
      "eval_loss": 1.0934925079345703,
      "eval_runtime": 413.7386,
      "eval_samples_per_second": 6.018,
      "eval_steps_per_second": 0.377,
      "step": 500
    },
    {
      "epoch": 0.04074315514993481,
      "grad_norm": 5.576170444488525,
      "learning_rate": 4.62962962962963e-05,
      "loss": 1.0336,
      "step": 1000
    },
    {
      "epoch": 0.04074315514993481,
      "eval_accuracy": 0.6514056224899598,
      "eval_f1": 0.6496455979119086,
      "eval_loss": 0.8499080538749695,
      "eval_runtime": 411.5812,
      "eval_samples_per_second": 6.05,
      "eval_steps_per_second": 0.379,
      "step": 1000
    },
    {
      "epoch": 0.06111473272490222,
      "grad_norm": 7.646302700042725,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.8636,
      "step": 1500
    },
    {
      "epoch": 0.06111473272490222,
      "eval_accuracy": 0.7269076305220884,
      "eval_f1": 0.7287323684689043,
      "eval_loss": 0.668684184551239,
      "eval_runtime": 411.6805,
      "eval_samples_per_second": 6.048,
      "eval_steps_per_second": 0.379,
      "step": 1500
    },
    {
      "epoch": 0.08148631029986962,
      "grad_norm": 11.417831420898438,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.7968,
      "step": 2000
    },
    {
      "epoch": 0.08148631029986962,
      "eval_accuracy": 0.734136546184739,
      "eval_f1": 0.7373689502738993,
      "eval_loss": 0.6560496091842651,
      "eval_runtime": 411.618,
      "eval_samples_per_second": 6.049,
      "eval_steps_per_second": 0.379,
      "step": 2000
    },
    {
      "epoch": 0.10185788787483703,
      "grad_norm": 5.362828254699707,
      "learning_rate": 3.240740740740741e-05,
      "loss": 0.7744,
      "step": 2500
    },
    {
      "epoch": 0.10185788787483703,
      "eval_accuracy": 0.7546184738955823,
      "eval_f1": 0.7561677733171569,
      "eval_loss": 0.6078478693962097,
      "eval_runtime": 411.4482,
      "eval_samples_per_second": 6.052,
      "eval_steps_per_second": 0.379,
      "step": 2500
    },
    {
      "epoch": 0.12222946544980444,
      "grad_norm": 5.366226673126221,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.7559,
      "step": 3000
    },
    {
      "epoch": 0.12222946544980444,
      "eval_accuracy": 0.7570281124497992,
      "eval_f1": 0.7592593531306621,
      "eval_loss": 0.6063372492790222,
      "eval_runtime": 411.1805,
      "eval_samples_per_second": 6.056,
      "eval_steps_per_second": 0.379,
      "step": 3000
    },
    {
      "epoch": 0.14260104302477183,
      "grad_norm": 7.356509685516357,
      "learning_rate": 2.314814814814815e-05,
      "loss": 0.7419,
      "step": 3500
    },
    {
      "epoch": 0.14260104302477183,
      "eval_accuracy": 0.7734939759036145,
      "eval_f1": 0.7749800378134201,
      "eval_loss": 0.5820600390434265,
      "eval_runtime": 411.2963,
      "eval_samples_per_second": 6.054,
      "eval_steps_per_second": 0.379,
      "step": 3500
    },
    {
      "epoch": 0.16297262059973924,
      "grad_norm": 6.393350124359131,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.7382,
      "step": 4000
    },
    {
      "epoch": 0.16297262059973924,
      "eval_accuracy": 0.7746987951807229,
      "eval_f1": 0.7750757863908069,
      "eval_loss": 0.5675287842750549,
      "eval_runtime": 411.5669,
      "eval_samples_per_second": 6.05,
      "eval_steps_per_second": 0.379,
      "step": 4000
    },
    {
      "epoch": 0.18334419817470665,
      "grad_norm": 6.975522994995117,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.7207,
      "step": 4500
    },
    {
      "epoch": 0.18334419817470665,
      "eval_accuracy": 0.7738955823293173,
      "eval_f1": 0.7758231525059284,
      "eval_loss": 0.5766967535018921,
      "eval_runtime": 411.2036,
      "eval_samples_per_second": 6.055,
      "eval_steps_per_second": 0.379,
      "step": 4500
    },
    {
      "epoch": 0.20371577574967406,
      "grad_norm": 6.244735240936279,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.7221,
      "step": 5000
    },
    {
      "epoch": 0.20371577574967406,
      "eval_accuracy": 0.7819277108433735,
      "eval_f1": 0.7828326523350859,
      "eval_loss": 0.5711483359336853,
      "eval_runtime": 432.5918,
      "eval_samples_per_second": 5.756,
      "eval_steps_per_second": 0.361,
      "step": 5000
    },
    {
      "epoch": 0.22408735332464147,
      "grad_norm": 4.810456275939941,
      "learning_rate": 4.6296296296296296e-06,
      "loss": 0.7079,
      "step": 5500
    },
    {
      "epoch": 0.22408735332464147,
      "eval_accuracy": 0.7823293172690763,
      "eval_f1": 0.783225399873768,
      "eval_loss": 0.5653966665267944,
      "eval_runtime": 411.5268,
      "eval_samples_per_second": 6.051,
      "eval_steps_per_second": 0.379,
      "step": 5500
    },
    {
      "epoch": 0.24445893089960888,
      "grad_norm": 7.606008052825928,
      "learning_rate": 0.0,
      "loss": 0.7073,
      "step": 6000
    },
    {
      "epoch": 0.24445893089960888,
      "eval_accuracy": 0.7835341365461848,
      "eval_f1": 0.7843248776628492,
      "eval_loss": 0.5673787593841553,
      "eval_runtime": 411.0953,
      "eval_samples_per_second": 6.057,
      "eval_steps_per_second": 0.379,
      "step": 6000
    }
  ],
  "logging_steps": 500,
  "max_steps": 6000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 8.03166870528e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}