last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04678362573099415,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004678362573099415,
      "grad_norm": 2.3281688690185547,
      "learning_rate": 2e-05,
      "loss": 5.789,
      "step": 1
    },
    {
      "epoch": 0.004678362573099415,
      "eval_loss": 1.711495280265808,
      "eval_runtime": 3.7612,
      "eval_samples_per_second": 23.928,
      "eval_steps_per_second": 11.964,
      "step": 1
    },
    {
      "epoch": 0.00935672514619883,
      "grad_norm": 2.8541603088378906,
      "learning_rate": 4e-05,
      "loss": 6.7761,
      "step": 2
    },
    {
      "epoch": 0.014035087719298246,
      "grad_norm": 4.04490852355957,
      "learning_rate": 6e-05,
      "loss": 8.094,
      "step": 3
    },
    {
      "epoch": 0.014035087719298246,
      "eval_loss": 1.7029690742492676,
      "eval_runtime": 2.37,
      "eval_samples_per_second": 37.975,
      "eval_steps_per_second": 18.987,
      "step": 3
    },
    {
      "epoch": 0.01871345029239766,
      "grad_norm": 2.8079593181610107,
      "learning_rate": 8e-05,
      "loss": 7.8941,
      "step": 4
    },
    {
      "epoch": 0.023391812865497075,
      "grad_norm": 4.025813102722168,
      "learning_rate": 0.0001,
      "loss": 9.5078,
      "step": 5
    },
    {
      "epoch": 0.028070175438596492,
      "grad_norm": 2.5388710498809814,
      "learning_rate": 0.00012,
      "loss": 6.1025,
      "step": 6
    },
    {
      "epoch": 0.028070175438596492,
      "eval_loss": 1.649599552154541,
      "eval_runtime": 2.3775,
      "eval_samples_per_second": 37.855,
      "eval_steps_per_second": 18.927,
      "step": 6
    },
    {
      "epoch": 0.03274853801169591,
      "grad_norm": 3.2335305213928223,
      "learning_rate": 0.00014,
      "loss": 7.23,
      "step": 7
    },
    {
      "epoch": 0.03742690058479532,
      "grad_norm": 3.0785489082336426,
      "learning_rate": 0.00016,
      "loss": 6.6232,
      "step": 8
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": 3.4422214031219482,
      "learning_rate": 0.00018,
      "loss": 7.527,
      "step": 9
    },
    {
      "epoch": 0.042105263157894736,
      "eval_loss": 1.5343419313430786,
      "eval_runtime": 2.3778,
      "eval_samples_per_second": 37.85,
      "eval_steps_per_second": 18.925,
      "step": 9
    },
    {
      "epoch": 0.04678362573099415,
      "grad_norm": 2.130618095397949,
      "learning_rate": 0.0002,
      "loss": 5.0657,
      "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": 1077366384230400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}