last-checkpoint/trainer_state.json

{
  "best_metric": 0.0076544685289263725,
  "best_model_checkpoint": "flan-search-terms/checkpoint-7500",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 7500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.33,
      "learning_rate": 4.666666666666667e-05,
      "loss": 3.6415,
      "step": 500
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 0.0455,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "learning_rate": 4e-05,
      "loss": 0.0301,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 4.940333333333333,
      "eval_loss": 0.01884176954627037,
      "eval_rouge1": 6.3132,
      "eval_rouge2": 5.5233,
      "eval_rougeL": 5.9306,
      "eval_rougeLsum": 5.9285,
      "eval_runtime": 220.0768,
      "eval_samples_per_second": 13.632,
      "eval_steps_per_second": 1.704,
      "step": 1500
    },
    {
      "epoch": 1.33,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 0.0238,
      "step": 2000
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0198,
      "step": 2500
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.0167,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 7.261666666666667,
      "eval_loss": 0.011832883581519127,
      "eval_rouge1": 5.6248,
      "eval_rouge2": 4.8018,
      "eval_rougeL": 5.2386,
      "eval_rougeLsum": 5.238,
      "eval_runtime": 232.4413,
      "eval_samples_per_second": 12.906,
      "eval_steps_per_second": 1.613,
      "step": 3000
    },
    {
      "epoch": 2.33,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.0152,
      "step": 3500
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 0.0143,
      "step": 4000
    },
    {
      "epoch": 3.0,
      "learning_rate": 2e-05,
      "loss": 0.0136,
      "step": 4500
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 8.599,
      "eval_loss": 0.009255478158593178,
      "eval_rouge1": 5.2543,
      "eval_rouge2": 4.4034,
      "eval_rougeL": 4.8643,
      "eval_rougeLsum": 4.8675,
      "eval_runtime": 252.1869,
      "eval_samples_per_second": 11.896,
      "eval_steps_per_second": 1.487,
      "step": 4500
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.0127,
      "step": 5000
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.0119,
      "step": 5500
    },
    {
      "epoch": 4.0,
      "learning_rate": 1e-05,
      "loss": 0.0114,
      "step": 6000
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 8.544666666666666,
      "eval_loss": 0.007984320633113384,
      "eval_rouge1": 5.3013,
      "eval_rouge2": 4.448,
      "eval_rougeL": 4.9117,
      "eval_rougeLsum": 4.9183,
      "eval_runtime": 255.3664,
      "eval_samples_per_second": 11.748,
      "eval_steps_per_second": 1.468,
      "step": 6000
    },
    {
      "epoch": 4.33,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.011,
      "step": 6500
    },
    {
      "epoch": 4.67,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.0112,
      "step": 7000
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.0105,
      "step": 7500
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 8.819,
      "eval_loss": 0.0076544685289263725,
      "eval_rouge1": 5.2275,
      "eval_rouge2": 4.3657,
      "eval_rougeL": 4.8365,
      "eval_rougeLsum": 4.8414,
      "eval_runtime": 257.9989,
      "eval_samples_per_second": 11.628,
      "eval_steps_per_second": 1.453,
      "step": 7500
    }
  ],
  "logging_steps": 500,
  "max_steps": 7500,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.115343028224e+16,
  "trial_name": null,
  "trial_params": null
}