last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0056409533211112676,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005640953321111268,
      "grad_norm": 3.7317214012145996,
      "learning_rate": 2e-05,
      "loss": 13.8478,
      "step": 1
    },
    {
      "epoch": 0.0005640953321111268,
      "eval_loss": 3.367891550064087,
      "eval_runtime": 50.8059,
      "eval_samples_per_second": 14.703,
      "eval_steps_per_second": 7.361,
      "step": 1
    },
    {
      "epoch": 0.0011281906642222536,
      "grad_norm": 5.836576461791992,
      "learning_rate": 4e-05,
      "loss": 14.5796,
      "step": 2
    },
    {
      "epoch": 0.0016922859963333804,
      "grad_norm": 4.388111591339111,
      "learning_rate": 6e-05,
      "loss": 14.3309,
      "step": 3
    },
    {
      "epoch": 0.0016922859963333804,
      "eval_loss": 3.3629956245422363,
      "eval_runtime": 50.7831,
      "eval_samples_per_second": 14.71,
      "eval_steps_per_second": 7.365,
      "step": 3
    },
    {
      "epoch": 0.0022563813284445073,
      "grad_norm": 3.77909517288208,
      "learning_rate": 8e-05,
      "loss": 13.6474,
      "step": 4
    },
    {
      "epoch": 0.0028204766605556338,
      "grad_norm": 3.8952269554138184,
      "learning_rate": 0.0001,
      "loss": 12.1901,
      "step": 5
    },
    {
      "epoch": 0.0033845719926667607,
      "grad_norm": 5.38345193862915,
      "learning_rate": 0.00012,
      "loss": 13.1139,
      "step": 6
    },
    {
      "epoch": 0.0033845719926667607,
      "eval_loss": 3.3215086460113525,
      "eval_runtime": 50.9782,
      "eval_samples_per_second": 14.653,
      "eval_steps_per_second": 7.336,
      "step": 6
    },
    {
      "epoch": 0.003948667324777887,
      "grad_norm": 7.791871070861816,
      "learning_rate": 0.00014,
      "loss": 14.1255,
      "step": 7
    },
    {
      "epoch": 0.0045127626568890146,
      "grad_norm": 4.868691921234131,
      "learning_rate": 0.00016,
      "loss": 13.5848,
      "step": 8
    },
    {
      "epoch": 0.005076857989000141,
      "grad_norm": 8.712459564208984,
      "learning_rate": 0.00018,
      "loss": 14.2394,
      "step": 9
    },
    {
      "epoch": 0.005076857989000141,
      "eval_loss": 3.1697006225585938,
      "eval_runtime": 50.9731,
      "eval_samples_per_second": 14.655,
      "eval_steps_per_second": 7.337,
      "step": 9
    },
    {
      "epoch": 0.0056409533211112676,
      "grad_norm": 7.540809631347656,
      "learning_rate": 0.0002,
      "loss": 12.9479,
      "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": 1628733135912960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}