last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.01412928293889085,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001412928293889085,
      "grad_norm": 1.074696660041809,
      "learning_rate": 2e-05,
      "loss": 44.2813,
      "step": 1
    },
    {
      "epoch": 0.001412928293889085,
      "eval_loss": 11.076643943786621,
      "eval_runtime": 1.0111,
      "eval_samples_per_second": 294.74,
      "eval_steps_per_second": 147.37,
      "step": 1
    },
    {
      "epoch": 0.00282585658777817,
      "grad_norm": 0.9785131216049194,
      "learning_rate": 4e-05,
      "loss": 44.3553,
      "step": 2
    },
    {
      "epoch": 0.0042387848816672555,
      "grad_norm": 1.0982869863510132,
      "learning_rate": 6e-05,
      "loss": 44.2961,
      "step": 3
    },
    {
      "epoch": 0.0042387848816672555,
      "eval_loss": 11.076664924621582,
      "eval_runtime": 0.6018,
      "eval_samples_per_second": 495.14,
      "eval_steps_per_second": 247.57,
      "step": 3
    },
    {
      "epoch": 0.00565171317555634,
      "grad_norm": 1.0487138032913208,
      "learning_rate": 8e-05,
      "loss": 44.3014,
      "step": 4
    },
    {
      "epoch": 0.007064641469445425,
      "grad_norm": 0.9543783664703369,
      "learning_rate": 0.0001,
      "loss": 44.3589,
      "step": 5
    },
    {
      "epoch": 0.008477569763334511,
      "grad_norm": 1.205977439880371,
      "learning_rate": 0.00012,
      "loss": 44.2548,
      "step": 6
    },
    {
      "epoch": 0.008477569763334511,
      "eval_loss": 11.07591438293457,
      "eval_runtime": 0.6104,
      "eval_samples_per_second": 488.231,
      "eval_steps_per_second": 244.115,
      "step": 6
    },
    {
      "epoch": 0.009890498057223597,
      "grad_norm": 1.1370278596878052,
      "learning_rate": 0.00014,
      "loss": 44.2413,
      "step": 7
    },
    {
      "epoch": 0.01130342635111268,
      "grad_norm": 1.2059990167617798,
      "learning_rate": 0.00016,
      "loss": 44.2927,
      "step": 8
    },
    {
      "epoch": 0.012716354645001766,
      "grad_norm": 1.1243544816970825,
      "learning_rate": 0.00018,
      "loss": 44.3321,
      "step": 9
    },
    {
      "epoch": 0.012716354645001766,
      "eval_loss": 11.074782371520996,
      "eval_runtime": 0.6157,
      "eval_samples_per_second": 483.972,
      "eval_steps_per_second": 241.986,
      "step": 9
    },
    {
      "epoch": 0.01412928293889085,
      "grad_norm": 1.0515074729919434,
      "learning_rate": 0.0002,
      "loss": 44.3293,
      "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": 7471104000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}