last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0004655981189835993,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.655981189835993e-05,
      "grad_norm": 1.3801172971725464,
      "learning_rate": 2e-05,
      "loss": 0.7699,
      "step": 1
    },
    {
      "epoch": 4.655981189835993e-05,
      "eval_loss": 1.0855792760849,
      "eval_runtime": 379.7475,
      "eval_samples_per_second": 23.816,
      "eval_steps_per_second": 11.908,
      "step": 1
    },
    {
      "epoch": 9.311962379671986e-05,
      "grad_norm": 1.1139503717422485,
      "learning_rate": 4e-05,
      "loss": 0.8808,
      "step": 2
    },
    {
      "epoch": 0.0001396794356950798,
      "grad_norm": 1.137139916419983,
      "learning_rate": 6e-05,
      "loss": 0.988,
      "step": 3
    },
    {
      "epoch": 0.0001396794356950798,
      "eval_loss": 1.0738060474395752,
      "eval_runtime": 378.8548,
      "eval_samples_per_second": 23.872,
      "eval_steps_per_second": 11.936,
      "step": 3
    },
    {
      "epoch": 0.00018623924759343972,
      "grad_norm": 1.2903538942337036,
      "learning_rate": 8e-05,
      "loss": 1.0027,
      "step": 4
    },
    {
      "epoch": 0.00023279905949179965,
      "grad_norm": 1.1982836723327637,
      "learning_rate": 0.0001,
      "loss": 0.8253,
      "step": 5
    },
    {
      "epoch": 0.0002793588713901596,
      "grad_norm": 1.0264619588851929,
      "learning_rate": 0.00012,
      "loss": 0.8042,
      "step": 6
    },
    {
      "epoch": 0.0002793588713901596,
      "eval_loss": 0.9054593443870544,
      "eval_runtime": 378.8922,
      "eval_samples_per_second": 23.87,
      "eval_steps_per_second": 11.935,
      "step": 6
    },
    {
      "epoch": 0.00032591868328851953,
      "grad_norm": 0.9776747822761536,
      "learning_rate": 0.00014,
      "loss": 0.602,
      "step": 7
    },
    {
      "epoch": 0.00037247849518687944,
      "grad_norm": 1.3397825956344604,
      "learning_rate": 0.00016,
      "loss": 0.6938,
      "step": 8
    },
    {
      "epoch": 0.0004190383070852394,
      "grad_norm": 2.170086145401001,
      "learning_rate": 0.00018,
      "loss": 0.6669,
      "step": 9
    },
    {
      "epoch": 0.0004190383070852394,
      "eval_loss": 0.6435133218765259,
      "eval_runtime": 379.1632,
      "eval_samples_per_second": 23.853,
      "eval_steps_per_second": 11.926,
      "step": 9
    },
    {
      "epoch": 0.0004655981189835993,
      "grad_norm": 1.5686155557632446,
      "learning_rate": 0.0002,
      "loss": 0.6175,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2589389947404288.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}