trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "grad_norm": 18.01788389344521,
      "learning_rate": 5e-06,
      "loss": 1.1132,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.9493625164031982,
      "eval_runtime": 5.3921,
      "eval_samples_per_second": 8.902,
      "eval_steps_per_second": 0.371,
      "step": 4
    },
    {
      "epoch": 1.25,
      "grad_norm": 9.96328065072019,
      "learning_rate": 9.847001329696653e-06,
      "loss": 1.0469,
      "step": 5
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5404321551322937,
      "eval_runtime": 5.3239,
      "eval_samples_per_second": 9.016,
      "eval_steps_per_second": 0.376,
      "step": 8
    },
    {
      "epoch": 2.5,
      "grad_norm": 6.243012201796606,
      "learning_rate": 8.94570254698197e-06,
      "loss": 0.6437,
      "step": 10
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.25905123353004456,
      "eval_runtime": 5.3216,
      "eval_samples_per_second": 9.02,
      "eval_steps_per_second": 0.376,
      "step": 12
    },
    {
      "epoch": 3.75,
      "grad_norm": 4.3348730787237155,
      "learning_rate": 7.379736965185369e-06,
      "loss": 0.2869,
      "step": 15
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.11114040017127991,
      "eval_runtime": 5.3277,
      "eval_samples_per_second": 9.009,
      "eval_steps_per_second": 0.375,
      "step": 16
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.31206970549863,
      "learning_rate": 5.412896727361663e-06,
      "loss": 0.1014,
      "step": 20
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.054268091917037964,
      "eval_runtime": 5.3319,
      "eval_samples_per_second": 9.002,
      "eval_steps_per_second": 0.375,
      "step": 20
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.03400101512670517,
      "eval_runtime": 5.3236,
      "eval_samples_per_second": 9.016,
      "eval_steps_per_second": 0.376,
      "step": 24
    },
    {
      "epoch": 6.25,
      "grad_norm": 0.9780308547160452,
      "learning_rate": 3.3765026539765832e-06,
      "loss": 0.046,
      "step": 25
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.022755322977900505,
      "eval_runtime": 5.3198,
      "eval_samples_per_second": 9.023,
      "eval_steps_per_second": 0.376,
      "step": 28
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.9473878282110842,
      "learning_rate": 1.6135921418712959e-06,
      "loss": 0.0282,
      "step": 30
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.01530417799949646,
      "eval_runtime": 5.3181,
      "eval_samples_per_second": 9.026,
      "eval_steps_per_second": 0.376,
      "step": 32
    },
    {
      "epoch": 8.75,
      "grad_norm": 0.5840687859842386,
      "learning_rate": 4.211333667247125e-07,
      "loss": 0.0157,
      "step": 35
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.011590150184929371,
      "eval_runtime": 5.3351,
      "eval_samples_per_second": 8.997,
      "eval_steps_per_second": 0.375,
      "step": 36
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.4445302982197704,
      "learning_rate": 0.0,
      "loss": 0.0111,
      "step": 40
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.010843921452760696,
      "eval_runtime": 5.2924,
      "eval_samples_per_second": 9.07,
      "eval_steps_per_second": 0.378,
      "step": 40
    },
    {
      "epoch": 10.0,
      "step": 40,
      "total_flos": 8375186227200.0,
      "train_loss": 0.2741479417309165,
      "train_runtime": 558.26,
      "train_samples_per_second": 1.773,
      "train_steps_per_second": 0.072
    }
  ],
  "logging_steps": 5,
  "max_steps": 40,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8375186227200.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}