last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.031931878658861094,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0010643959552953698,
      "eval_loss": NaN,
      "eval_runtime": 49.8875,
      "eval_samples_per_second": 7.938,
      "eval_steps_per_second": 3.969,
      "step": 1
    },
    {
      "epoch": 0.0031931878658861094,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 5.9521,
      "step": 3
    },
    {
      "epoch": 0.005321979776476849,
      "eval_loss": NaN,
      "eval_runtime": 49.8334,
      "eval_samples_per_second": 7.946,
      "eval_steps_per_second": 3.973,
      "step": 5
    },
    {
      "epoch": 0.006386375731772219,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.009579563597658328,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.010643959552953698,
      "eval_loss": NaN,
      "eval_runtime": 49.9038,
      "eval_samples_per_second": 7.935,
      "eval_steps_per_second": 3.968,
      "step": 10
    },
    {
      "epoch": 0.012772751463544438,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.015965939329430547,
      "grad_norm": NaN,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.015965939329430547,
      "eval_loss": NaN,
      "eval_runtime": 49.9074,
      "eval_samples_per_second": 7.935,
      "eval_steps_per_second": 3.967,
      "step": 15
    },
    {
      "epoch": 0.019159127195316657,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.021287919105907396,
      "eval_loss": NaN,
      "eval_runtime": 49.9305,
      "eval_samples_per_second": 7.931,
      "eval_steps_per_second": 3.966,
      "step": 20
    },
    {
      "epoch": 0.022352315061202766,
      "grad_norm": NaN,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.025545502927088876,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.026609898882384245,
      "eval_loss": NaN,
      "eval_runtime": 49.8871,
      "eval_samples_per_second": 7.938,
      "eval_steps_per_second": 3.969,
      "step": 25
    },
    {
      "epoch": 0.028738690792974985,
      "grad_norm": NaN,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.031931878658861094,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.031931878658861094,
      "eval_loss": NaN,
      "eval_runtime": 49.9111,
      "eval_samples_per_second": 7.934,
      "eval_steps_per_second": 3.967,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4803529675898880.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}