trainer_state.json

{
  "best_metric": 0.06724460422992706,
  "best_model_checkpoint": "/data/jcanete/all_results/pos/albeto_xlarge/epochs_4_bs_32_lr_5e-6/checkpoint-1600",
  "epoch": 4.0,
  "global_step": 1792,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.45,
      "eval_accuracy": 0.9701462191423391,
      "eval_f1": 0.9659476055664102,
      "eval_loss": 0.11576754599809647,
      "eval_precision": 0.9638913757087436,
      "eval_recall": 0.9680126271203373,
      "eval_runtime": 8.2971,
      "eval_samples_per_second": 199.347,
      "eval_steps_per_second": 6.267,
      "step": 200
    },
    {
      "epoch": 0.89,
      "eval_accuracy": 0.9774148184544484,
      "eval_f1": 0.9743559053233191,
      "eval_loss": 0.08359552174806595,
      "eval_precision": 0.9728518503764491,
      "eval_recall": 0.9758646180895486,
      "eval_runtime": 8.2688,
      "eval_samples_per_second": 200.028,
      "eval_steps_per_second": 6.289,
      "step": 400
    },
    {
      "epoch": 1.12,
      "learning_rate": 3.61328125e-06,
      "loss": 0.3126,
      "step": 500
    },
    {
      "epoch": 1.34,
      "eval_accuracy": 0.9796174243066027,
      "eval_f1": 0.9774613052497209,
      "eval_loss": 0.07587467133998871,
      "eval_precision": 0.9758069333545728,
      "eval_recall": 0.9791212962777966,
      "eval_runtime": 8.2627,
      "eval_samples_per_second": 200.176,
      "eval_steps_per_second": 6.293,
      "step": 600
    },
    {
      "epoch": 1.79,
      "eval_accuracy": 0.981243964012809,
      "eval_f1": 0.9795327764971211,
      "eval_loss": 0.07019403576850891,
      "eval_precision": 0.978487977989553,
      "eval_recall": 0.9805798085952329,
      "eval_runtime": 8.2592,
      "eval_samples_per_second": 200.262,
      "eval_steps_per_second": 6.296,
      "step": 800
    },
    {
      "epoch": 2.23,
      "learning_rate": 2.2181919642857142e-06,
      "loss": 0.0602,
      "step": 1000
    },
    {
      "epoch": 2.23,
      "eval_accuracy": 0.9813286796225072,
      "eval_f1": 0.9797755051134948,
      "eval_loss": 0.06836934387683868,
      "eval_precision": 0.9785745605293579,
      "eval_recall": 0.9809794010109688,
      "eval_runtime": 8.2691,
      "eval_samples_per_second": 200.021,
      "eval_steps_per_second": 6.288,
      "step": 1000
    },
    {
      "epoch": 2.68,
      "eval_accuracy": 0.981531997085783,
      "eval_f1": 0.9803721997704935,
      "eval_loss": 0.06962081789970398,
      "eval_precision": 0.9792680009568615,
      "eval_recall": 0.9814788915306387,
      "eval_runtime": 8.2695,
      "eval_samples_per_second": 200.012,
      "eval_steps_per_second": 6.288,
      "step": 1200
    },
    {
      "epoch": 3.12,
      "eval_accuracy": 0.9818369732806967,
      "eval_f1": 0.9806183756162798,
      "eval_loss": 0.06813845783472061,
      "eval_precision": 0.9796797415598142,
      "eval_recall": 0.981558810013786,
      "eval_runtime": 8.2599,
      "eval_samples_per_second": 200.245,
      "eval_steps_per_second": 6.295,
      "step": 1400
    },
    {
      "epoch": 3.35,
      "learning_rate": 8.231026785714287e-07,
      "loss": 0.0439,
      "step": 1500
    },
    {
      "epoch": 3.57,
      "eval_accuracy": 0.9816844851832399,
      "eval_f1": 0.980655593707578,
      "eval_loss": 0.06724460422992706,
      "eval_precision": 0.9798535894520575,
      "eval_recall": 0.9814589119098519,
      "eval_runtime": 8.2557,
      "eval_samples_per_second": 200.345,
      "eval_steps_per_second": 6.299,
      "step": 1600
    },
    {
      "epoch": 4.0,
      "step": 1792,
      "total_flos": 1487933302262400.0,
      "train_loss": 0.12245277049286026,
      "train_runtime": 1186.9763,
      "train_samples_per_second": 48.207,
      "train_steps_per_second": 1.51
    }
  ],
  "max_steps": 1792,
  "num_train_epochs": 4,
  "total_flos": 1487933302262400.0,
  "trial_name": null,
  "trial_params": null
}