last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005115816399033679,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005684240443370755,
      "grad_norm": 1.2595114707946777,
      "learning_rate": 2e-05,
      "loss": 2.0859,
      "step": 1
    },
    {
      "epoch": 0.0005684240443370755,
      "eval_loss": 2.5626769065856934,
      "eval_runtime": 26.5397,
      "eval_samples_per_second": 27.92,
      "eval_steps_per_second": 13.979,
      "step": 1
    },
    {
      "epoch": 0.001136848088674151,
      "grad_norm": 1.450069546699524,
      "learning_rate": 4e-05,
      "loss": 2.3014,
      "step": 2
    },
    {
      "epoch": 0.0017052721330112265,
      "grad_norm": 1.2272709608078003,
      "learning_rate": 6e-05,
      "loss": 2.3704,
      "step": 3
    },
    {
      "epoch": 0.0017052721330112265,
      "eval_loss": 2.556915044784546,
      "eval_runtime": 24.7613,
      "eval_samples_per_second": 29.926,
      "eval_steps_per_second": 14.983,
      "step": 3
    },
    {
      "epoch": 0.002273696177348302,
      "grad_norm": 1.4716559648513794,
      "learning_rate": 8e-05,
      "loss": 2.8779,
      "step": 4
    },
    {
      "epoch": 0.002842120221685377,
      "grad_norm": 1.7448276281356812,
      "learning_rate": 0.0001,
      "loss": 2.546,
      "step": 5
    },
    {
      "epoch": 0.003410544266022453,
      "grad_norm": 1.155127763748169,
      "learning_rate": 0.00012,
      "loss": 2.2414,
      "step": 6
    },
    {
      "epoch": 0.003410544266022453,
      "eval_loss": 2.4394571781158447,
      "eval_runtime": 24.8085,
      "eval_samples_per_second": 29.869,
      "eval_steps_per_second": 14.955,
      "step": 6
    },
    {
      "epoch": 0.003978968310359528,
      "grad_norm": 1.2924119234085083,
      "learning_rate": 0.00014,
      "loss": 2.1171,
      "step": 7
    },
    {
      "epoch": 0.004547392354696604,
      "grad_norm": 1.4125940799713135,
      "learning_rate": 0.00016,
      "loss": 2.2975,
      "step": 8
    },
    {
      "epoch": 0.005115816399033679,
      "grad_norm": 1.473420262336731,
      "learning_rate": 0.00018,
      "loss": 2.2776,
      "step": 9
    },
    {
      "epoch": 0.005115816399033679,
      "eval_loss": 2.213955879211426,
      "eval_runtime": 24.7812,
      "eval_samples_per_second": 29.902,
      "eval_steps_per_second": 14.971,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1710846929534976.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}