trainer_state.json

{
  "best_metric": 0.9879474725670084,
  "best_model_checkpoint": "pasha/checkpoint-500",
  "epoch": 21.27659574468085,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.13,
      "eval_accuracy": 0.9570877531340405,
      "eval_f1": 0.9485879797248371,
      "eval_loss": 0.2664182484149933,
      "eval_precision": 0.9534206695778749,
      "eval_recall": 0.9438040345821326,
      "eval_runtime": 13.9899,
      "eval_samples_per_second": 13.51,
      "eval_steps_per_second": 0.858,
      "step": 100
    },
    {
      "epoch": 4.26,
      "eval_accuracy": 0.983847637415622,
      "eval_f1": 0.9778975741239893,
      "eval_loss": 0.10435084253549576,
      "eval_precision": 0.9756185012549301,
      "eval_recall": 0.9801873198847262,
      "eval_runtime": 13.823,
      "eval_samples_per_second": 13.673,
      "eval_steps_per_second": 0.868,
      "step": 200
    },
    {
      "epoch": 6.38,
      "eval_accuracy": 0.9903567984570878,
      "eval_f1": 0.987601078167116,
      "eval_loss": 0.06718672811985016,
      "eval_precision": 0.9852993904625313,
      "eval_recall": 0.9899135446685879,
      "eval_runtime": 13.9839,
      "eval_samples_per_second": 13.516,
      "eval_steps_per_second": 0.858,
      "step": 300
    },
    {
      "epoch": 8.51,
      "eval_accuracy": 0.9884281581485053,
      "eval_f1": 0.9841783531103919,
      "eval_loss": 0.06342343986034393,
      "eval_precision": 0.9824120603015075,
      "eval_recall": 0.9859510086455331,
      "eval_runtime": 13.8826,
      "eval_samples_per_second": 13.614,
      "eval_steps_per_second": 0.864,
      "step": 400
    },
    {
      "epoch": 10.64,
      "learning_rate": 5e-06,
      "loss": 0.2958,
      "step": 500
    },
    {
      "epoch": 10.64,
      "eval_accuracy": 0.9905978784956606,
      "eval_f1": 0.9879474725670084,
      "eval_loss": 0.058533914387226105,
      "eval_precision": 0.986704994610133,
      "eval_recall": 0.989193083573487,
      "eval_runtime": 13.739,
      "eval_samples_per_second": 13.756,
      "eval_steps_per_second": 0.873,
      "step": 500
    },
    {
      "epoch": 12.77,
      "eval_accuracy": 0.9927675988428158,
      "eval_f1": 0.9908322847384505,
      "eval_loss": 0.051136456429958344,
      "eval_precision": 0.9888769285970578,
      "eval_recall": 0.9927953890489913,
      "eval_runtime": 14.0393,
      "eval_samples_per_second": 13.462,
      "eval_steps_per_second": 0.855,
      "step": 600
    },
    {
      "epoch": 14.89,
      "eval_accuracy": 0.992526518804243,
      "eval_f1": 0.9895795903701042,
      "eval_loss": 0.05025022476911545,
      "eval_precision": 0.9870967741935484,
      "eval_recall": 0.9920749279538905,
      "eval_runtime": 14.0132,
      "eval_samples_per_second": 13.487,
      "eval_steps_per_second": 0.856,
      "step": 700
    },
    {
      "epoch": 17.02,
      "eval_accuracy": 0.991321118611379,
      "eval_f1": 0.9881380301941048,
      "eval_loss": 0.05291323363780975,
      "eval_precision": 0.9860114777618364,
      "eval_recall": 0.9902737752161384,
      "eval_runtime": 13.9493,
      "eval_samples_per_second": 13.549,
      "eval_steps_per_second": 0.86,
      "step": 800
    },
    {
      "epoch": 19.15,
      "eval_accuracy": 0.9903567984570878,
      "eval_f1": 0.986704994610133,
      "eval_loss": 0.0581122450530529,
      "eval_precision": 0.9842293906810036,
      "eval_recall": 0.989193083573487,
      "eval_runtime": 13.9545,
      "eval_samples_per_second": 13.544,
      "eval_steps_per_second": 0.86,
      "step": 900
    },
    {
      "epoch": 21.28,
      "learning_rate": 0.0,
      "loss": 0.0256,
      "step": 1000
    },
    {
      "epoch": 21.28,
      "eval_accuracy": 0.990115718418515,
      "eval_f1": 0.9868775840373899,
      "eval_loss": 0.057054802775382996,
      "eval_precision": 0.984930032292788,
      "eval_recall": 0.9888328530259366,
      "eval_runtime": 14.2093,
      "eval_samples_per_second": 13.301,
      "eval_steps_per_second": 0.845,
      "step": 1000
    },
    {
      "epoch": 21.28,
      "step": 1000,
      "total_flos": 4247054450688000.0,
      "train_loss": 0.16068801975250244,
      "train_runtime": 1757.2853,
      "train_samples_per_second": 9.105,
      "train_steps_per_second": 0.569
    }
  ],
  "max_steps": 1000,
  "num_train_epochs": 22,
  "total_flos": 4247054450688000.0,
  "trial_name": null,
  "trial_params": null
}