trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3970,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 0.0083,
      "eval_gen_len": 19.0,
      "eval_loss": 1.1680346727371216,
      "eval_runtime": 38.281,
      "eval_samples_per_second": 20.689,
      "eval_steps_per_second": 1.306,
      "step": 397
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.370277078085643e-05,
      "loss": 1.7159,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.0165,
      "eval_gen_len": 19.0,
      "eval_loss": 1.0287196636199951,
      "eval_runtime": 38.488,
      "eval_samples_per_second": 20.578,
      "eval_steps_per_second": 1.299,
      "step": 794
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.7405541561712845e-05,
      "loss": 1.1916,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 0.0177,
      "eval_gen_len": 19.0,
      "eval_loss": 0.9632178544998169,
      "eval_runtime": 38.4087,
      "eval_samples_per_second": 20.62,
      "eval_steps_per_second": 1.302,
      "step": 1191
    },
    {
      "epoch": 3.78,
      "learning_rate": 3.1108312342569276e-05,
      "loss": 1.077,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.0307,
      "eval_gen_len": 19.0,
      "eval_loss": 0.9211758971214294,
      "eval_runtime": 38.4332,
      "eval_samples_per_second": 20.607,
      "eval_steps_per_second": 1.301,
      "step": 1588
    },
    {
      "epoch": 5.0,
      "eval_bleu": 0.0278,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8920512199401855,
      "eval_runtime": 39.0152,
      "eval_samples_per_second": 20.3,
      "eval_steps_per_second": 1.282,
      "step": 1985
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.4811083123425694e-05,
      "loss": 1.0159,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 0.0465,
      "eval_gen_len": 19.0,
      "eval_loss": 0.869423508644104,
      "eval_runtime": 38.5496,
      "eval_samples_per_second": 20.545,
      "eval_steps_per_second": 1.297,
      "step": 2382
    },
    {
      "epoch": 6.3,
      "learning_rate": 1.8513853904282116e-05,
      "loss": 0.9723,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 0.0489,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8542945981025696,
      "eval_runtime": 38.3887,
      "eval_samples_per_second": 20.631,
      "eval_steps_per_second": 1.302,
      "step": 2779
    },
    {
      "epoch": 7.56,
      "learning_rate": 1.2216624685138539e-05,
      "loss": 0.9457,
      "step": 3000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 0.0451,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8452380895614624,
      "eval_runtime": 38.2635,
      "eval_samples_per_second": 20.699,
      "eval_steps_per_second": 1.307,
      "step": 3176
    },
    {
      "epoch": 8.82,
      "learning_rate": 5.919395465994963e-06,
      "loss": 0.9276,
      "step": 3500
    },
    {
      "epoch": 9.0,
      "eval_bleu": 0.0496,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8380882143974304,
      "eval_runtime": 38.7379,
      "eval_samples_per_second": 20.445,
      "eval_steps_per_second": 1.291,
      "step": 3573
    },
    {
      "epoch": 10.0,
      "eval_bleu": 0.048,
      "eval_gen_len": 19.0,
      "eval_loss": 0.836793839931488,
      "eval_runtime": 38.47,
      "eval_samples_per_second": 20.587,
      "eval_steps_per_second": 1.3,
      "step": 3970
    },
    {
      "epoch": 10.0,
      "step": 3970,
      "total_flos": 2.2534682441342976e+16,
      "train_loss": 1.0967687035087375,
      "train_runtime": 2092.8002,
      "train_samples_per_second": 30.28,
      "train_steps_per_second": 1.897
    }
  ],
  "max_steps": 3970,
  "num_train_epochs": 10,
  "total_flos": 2.2534682441342976e+16,
  "trial_name": null,
  "trial_params": null
}