last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0003989706557082727,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.9897065570827266e-05,
      "grad_norm": 0.8825915455818176,
      "learning_rate": 2e-05,
      "loss": 1.968,
      "step": 1
    },
    {
      "epoch": 3.9897065570827266e-05,
      "eval_loss": 2.162811517715454,
      "eval_runtime": 597.3588,
      "eval_samples_per_second": 17.668,
      "eval_steps_per_second": 8.834,
      "step": 1
    },
    {
      "epoch": 7.979413114165453e-05,
      "grad_norm": 0.9215392470359802,
      "learning_rate": 4e-05,
      "loss": 2.0545,
      "step": 2
    },
    {
      "epoch": 0.0001196911967124818,
      "grad_norm": 0.5958425998687744,
      "learning_rate": 6e-05,
      "loss": 2.0149,
      "step": 3
    },
    {
      "epoch": 0.0001196911967124818,
      "eval_loss": 2.158785581588745,
      "eval_runtime": 597.9083,
      "eval_samples_per_second": 17.652,
      "eval_steps_per_second": 8.826,
      "step": 3
    },
    {
      "epoch": 0.00015958826228330906,
      "grad_norm": 0.9603627920150757,
      "learning_rate": 8e-05,
      "loss": 2.0924,
      "step": 4
    },
    {
      "epoch": 0.00019948532785413634,
      "grad_norm": 0.9111735224723816,
      "learning_rate": 0.0001,
      "loss": 2.2522,
      "step": 5
    },
    {
      "epoch": 0.0002393823934249636,
      "grad_norm": 0.8802748322486877,
      "learning_rate": 0.00012,
      "loss": 2.1143,
      "step": 6
    },
    {
      "epoch": 0.0002393823934249636,
      "eval_loss": 2.1063265800476074,
      "eval_runtime": 597.8051,
      "eval_samples_per_second": 17.655,
      "eval_steps_per_second": 8.827,
      "step": 6
    },
    {
      "epoch": 0.00027927945899579087,
      "grad_norm": 1.0103797912597656,
      "learning_rate": 0.00014,
      "loss": 1.9791,
      "step": 7
    },
    {
      "epoch": 0.0003191765245666181,
      "grad_norm": 0.7773309350013733,
      "learning_rate": 0.00016,
      "loss": 1.85,
      "step": 8
    },
    {
      "epoch": 0.0003590735901374454,
      "grad_norm": 1.2363407611846924,
      "learning_rate": 0.00018,
      "loss": 2.0996,
      "step": 9
    },
    {
      "epoch": 0.0003590735901374454,
      "eval_loss": 1.9484398365020752,
      "eval_runtime": 598.0406,
      "eval_samples_per_second": 17.648,
      "eval_steps_per_second": 8.824,
      "step": 9
    },
    {
      "epoch": 0.0003989706557082727,
      "grad_norm": 0.9276331067085266,
      "learning_rate": 0.0002,
      "loss": 1.9981,
      "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": 1578593680097280.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}