last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.2684563758389262,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.026845637583892617,
      "grad_norm": 32.97285079956055,
      "learning_rate": 2e-05,
      "loss": 10.7683,
      "step": 1
    },
    {
      "epoch": 0.026845637583892617,
      "eval_loss": 11.574974060058594,
      "eval_runtime": 4.5018,
      "eval_samples_per_second": 1.777,
      "eval_steps_per_second": 1.777,
      "step": 1
    },
    {
      "epoch": 0.053691275167785234,
      "grad_norm": 25.372859954833984,
      "learning_rate": 4e-05,
      "loss": 8.4845,
      "step": 2
    },
    {
      "epoch": 0.08053691275167785,
      "grad_norm": 30.13886070251465,
      "learning_rate": 6e-05,
      "loss": 11.9573,
      "step": 3
    },
    {
      "epoch": 0.08053691275167785,
      "eval_loss": 10.982644081115723,
      "eval_runtime": 4.4823,
      "eval_samples_per_second": 1.785,
      "eval_steps_per_second": 1.785,
      "step": 3
    },
    {
      "epoch": 0.10738255033557047,
      "grad_norm": 28.508464813232422,
      "learning_rate": 8e-05,
      "loss": 10.0268,
      "step": 4
    },
    {
      "epoch": 0.1342281879194631,
      "grad_norm": 48.71175765991211,
      "learning_rate": 0.0001,
      "loss": 10.3586,
      "step": 5
    },
    {
      "epoch": 0.1610738255033557,
      "grad_norm": 33.29041290283203,
      "learning_rate": 0.00012,
      "loss": 7.615,
      "step": 6
    },
    {
      "epoch": 0.1610738255033557,
      "eval_loss": 4.839656829833984,
      "eval_runtime": 4.4948,
      "eval_samples_per_second": 1.78,
      "eval_steps_per_second": 1.78,
      "step": 6
    },
    {
      "epoch": 0.18791946308724833,
      "grad_norm": 33.47844696044922,
      "learning_rate": 0.00014,
      "loss": 4.5034,
      "step": 7
    },
    {
      "epoch": 0.21476510067114093,
      "grad_norm": 18.189189910888672,
      "learning_rate": 0.00016,
      "loss": 2.4897,
      "step": 8
    },
    {
      "epoch": 0.24161073825503357,
      "grad_norm": 14.530669212341309,
      "learning_rate": 0.00018,
      "loss": 1.5332,
      "step": 9
    },
    {
      "epoch": 0.24161073825503357,
      "eval_loss": 1.0043643712997437,
      "eval_runtime": 4.4829,
      "eval_samples_per_second": 1.785,
      "eval_steps_per_second": 1.785,
      "step": 9
    },
    {
      "epoch": 0.2684563758389262,
      "grad_norm": 10.115413665771484,
      "learning_rate": 0.0002,
      "loss": 0.9297,
      "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": 4104787234652160.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}