last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00021398956800855958,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.139895680085596e-05,
      "grad_norm": 2.9232585430145264,
      "learning_rate": 2e-05,
      "loss": 4.6304,
      "step": 1
    },
    {
      "epoch": 2.139895680085596e-05,
      "eval_loss": 4.513030052185059,
      "eval_runtime": 609.7091,
      "eval_samples_per_second": 32.273,
      "eval_steps_per_second": 16.137,
      "step": 1
    },
    {
      "epoch": 4.279791360171192e-05,
      "grad_norm": 2.5452752113342285,
      "learning_rate": 4e-05,
      "loss": 4.6916,
      "step": 2
    },
    {
      "epoch": 6.419687040256787e-05,
      "grad_norm": 2.117677927017212,
      "learning_rate": 6e-05,
      "loss": 3.4051,
      "step": 3
    },
    {
      "epoch": 6.419687040256787e-05,
      "eval_loss": 4.505028247833252,
      "eval_runtime": 608.4975,
      "eval_samples_per_second": 32.337,
      "eval_steps_per_second": 16.169,
      "step": 3
    },
    {
      "epoch": 8.559582720342384e-05,
      "grad_norm": 2.447026252746582,
      "learning_rate": 8e-05,
      "loss": 4.2223,
      "step": 4
    },
    {
      "epoch": 0.00010699478400427979,
      "grad_norm": 2.3093831539154053,
      "learning_rate": 0.0001,
      "loss": 3.6027,
      "step": 5
    },
    {
      "epoch": 0.00012839374080513574,
      "grad_norm": 2.769754648208618,
      "learning_rate": 0.00012,
      "loss": 4.1278,
      "step": 6
    },
    {
      "epoch": 0.00012839374080513574,
      "eval_loss": 4.381775856018066,
      "eval_runtime": 609.2425,
      "eval_samples_per_second": 32.297,
      "eval_steps_per_second": 16.15,
      "step": 6
    },
    {
      "epoch": 0.0001497926976059917,
      "grad_norm": 2.932274341583252,
      "learning_rate": 0.00014,
      "loss": 4.2332,
      "step": 7
    },
    {
      "epoch": 0.00017119165440684768,
      "grad_norm": 3.290330410003662,
      "learning_rate": 0.00016,
      "loss": 4.3013,
      "step": 8
    },
    {
      "epoch": 0.00019259061120770361,
      "grad_norm": 4.83338737487793,
      "learning_rate": 0.00018,
      "loss": 4.0746,
      "step": 9
    },
    {
      "epoch": 0.00019259061120770361,
      "eval_loss": 3.6674344539642334,
      "eval_runtime": 607.6923,
      "eval_samples_per_second": 32.38,
      "eval_steps_per_second": 16.191,
      "step": 9
    },
    {
      "epoch": 0.00021398956800855958,
      "grad_norm": 4.905086040496826,
      "learning_rate": 0.0002,
      "loss": 3.2257,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}