last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02702702702702703,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002702702702702703,
      "grad_norm": 0.22516648471355438,
      "learning_rate": 2e-05,
      "loss": 2.4235,
      "step": 1
    },
    {
      "epoch": 0.002702702702702703,
      "eval_loss": 2.200819492340088,
      "eval_runtime": 8.0954,
      "eval_samples_per_second": 9.635,
      "eval_steps_per_second": 9.635,
      "step": 1
    },
    {
      "epoch": 0.005405405405405406,
      "grad_norm": 0.2626628577709198,
      "learning_rate": 4e-05,
      "loss": 2.2166,
      "step": 2
    },
    {
      "epoch": 0.008108108108108109,
      "grad_norm": 0.21446576714515686,
      "learning_rate": 6e-05,
      "loss": 2.6546,
      "step": 3
    },
    {
      "epoch": 0.008108108108108109,
      "eval_loss": 2.199615955352783,
      "eval_runtime": 8.2871,
      "eval_samples_per_second": 9.412,
      "eval_steps_per_second": 9.412,
      "step": 3
    },
    {
      "epoch": 0.010810810810810811,
      "grad_norm": 0.2510293424129486,
      "learning_rate": 8e-05,
      "loss": 2.6251,
      "step": 4
    },
    {
      "epoch": 0.013513513513513514,
      "grad_norm": 0.2600695788860321,
      "learning_rate": 0.0001,
      "loss": 2.0934,
      "step": 5
    },
    {
      "epoch": 0.016216216216216217,
      "grad_norm": 0.23503589630126953,
      "learning_rate": 0.00012,
      "loss": 2.3495,
      "step": 6
    },
    {
      "epoch": 0.016216216216216217,
      "eval_loss": 2.193514585494995,
      "eval_runtime": 8.4696,
      "eval_samples_per_second": 9.209,
      "eval_steps_per_second": 9.209,
      "step": 6
    },
    {
      "epoch": 0.01891891891891892,
      "grad_norm": 0.23834799230098724,
      "learning_rate": 0.00014,
      "loss": 2.0365,
      "step": 7
    },
    {
      "epoch": 0.021621621621621623,
      "grad_norm": 0.34777796268463135,
      "learning_rate": 0.00016,
      "loss": 2.0777,
      "step": 8
    },
    {
      "epoch": 0.024324324324324326,
      "grad_norm": 0.2863074839115143,
      "learning_rate": 0.00018,
      "loss": 2.3295,
      "step": 9
    },
    {
      "epoch": 0.024324324324324326,
      "eval_loss": 2.1840600967407227,
      "eval_runtime": 8.339,
      "eval_samples_per_second": 9.354,
      "eval_steps_per_second": 9.354,
      "step": 9
    },
    {
      "epoch": 0.02702702702702703,
      "grad_norm": 0.25496789813041687,
      "learning_rate": 0.0002,
      "loss": 2.1022,
      "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": 3467932965273600.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}