last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.20151133501259447,
  "eval_steps": 5,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010075566750629723,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.010075566750629723,
      "eval_loss": NaN,
      "eval_runtime": 94.6108,
      "eval_samples_per_second": 1.776,
      "eval_steps_per_second": 0.888,
      "step": 1
    },
    {
      "epoch": 0.020151133501259445,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.030226700251889168,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 2461.7207,
      "step": 3
    },
    {
      "epoch": 0.04030226700251889,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 64135.8789,
      "step": 4
    },
    {
      "epoch": 0.05037783375314862,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 225.9553,
      "step": 5
    },
    {
      "epoch": 0.05037783375314862,
      "eval_loss": NaN,
      "eval_runtime": 6.3005,
      "eval_samples_per_second": 26.664,
      "eval_steps_per_second": 13.332,
      "step": 5
    },
    {
      "epoch": 0.060453400503778336,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.07052896725440806,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.08060453400503778,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 8.8876,
      "step": 8
    },
    {
      "epoch": 0.0906801007556675,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 72.5001,
      "step": 9
    },
    {
      "epoch": 0.10075566750629723,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.10075566750629723,
      "eval_loss": NaN,
      "eval_runtime": 11.3956,
      "eval_samples_per_second": 14.743,
      "eval_steps_per_second": 7.371,
      "step": 10
    },
    {
      "epoch": 0.11083123425692695,
      "grad_norm": NaN,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.12090680100755667,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 11758.1523,
      "step": 12
    },
    {
      "epoch": 0.1309823677581864,
      "grad_norm": NaN,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.14105793450881612,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.15113350125944586,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.15113350125944586,
      "eval_loss": NaN,
      "eval_runtime": 10.5359,
      "eval_samples_per_second": 15.946,
      "eval_steps_per_second": 7.973,
      "step": 15
    },
    {
      "epoch": 0.16120906801007556,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.1712846347607053,
      "grad_norm": NaN,
      "learning_rate": 2.061073738537635e-05,
      "loss": 367.4662,
      "step": 17
    },
    {
      "epoch": 0.181360201511335,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.19143576826196473,
      "grad_norm": NaN,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 42088.5391,
      "step": 19
    },
    {
      "epoch": 0.20151133501259447,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 137961.625,
      "step": 20
    },
    {
      "epoch": 0.20151133501259447,
      "eval_loss": NaN,
      "eval_runtime": 11.3375,
      "eval_samples_per_second": 14.818,
      "eval_steps_per_second": 7.409,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.567664185848627e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}