last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.030534351145038167,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0030534351145038168,
      "grad_norm": 7.624812602996826,
      "learning_rate": 2e-05,
      "loss": 3.721,
      "step": 1
    },
    {
      "epoch": 0.0030534351145038168,
      "eval_loss": 2.613229513168335,
      "eval_runtime": 18.2389,
      "eval_samples_per_second": 7.566,
      "eval_steps_per_second": 3.783,
      "step": 1
    },
    {
      "epoch": 0.0061068702290076335,
      "grad_norm": 2.746663808822632,
      "learning_rate": 4e-05,
      "loss": 2.0161,
      "step": 2
    },
    {
      "epoch": 0.00916030534351145,
      "grad_norm": 1.5935509204864502,
      "learning_rate": 6e-05,
      "loss": 2.4389,
      "step": 3
    },
    {
      "epoch": 0.00916030534351145,
      "eval_loss": 2.592705726623535,
      "eval_runtime": 18.95,
      "eval_samples_per_second": 7.282,
      "eval_steps_per_second": 3.641,
      "step": 3
    },
    {
      "epoch": 0.012213740458015267,
      "grad_norm": 1.3037679195404053,
      "learning_rate": 8e-05,
      "loss": 2.2116,
      "step": 4
    },
    {
      "epoch": 0.015267175572519083,
      "grad_norm": 1.0085657835006714,
      "learning_rate": 0.0001,
      "loss": 1.9184,
      "step": 5
    },
    {
      "epoch": 0.0183206106870229,
      "grad_norm": 4.906698226928711,
      "learning_rate": 0.00012,
      "loss": 2.8942,
      "step": 6
    },
    {
      "epoch": 0.0183206106870229,
      "eval_loss": 2.3518502712249756,
      "eval_runtime": 19.535,
      "eval_samples_per_second": 7.064,
      "eval_steps_per_second": 3.532,
      "step": 6
    },
    {
      "epoch": 0.021374045801526718,
      "grad_norm": 6.759243965148926,
      "learning_rate": 0.00014,
      "loss": 2.655,
      "step": 7
    },
    {
      "epoch": 0.024427480916030534,
      "grad_norm": 0.7865249514579773,
      "learning_rate": 0.00016,
      "loss": 2.0324,
      "step": 8
    },
    {
      "epoch": 0.02748091603053435,
      "grad_norm": 1.0031616687774658,
      "learning_rate": 0.00018,
      "loss": 1.7614,
      "step": 9
    },
    {
      "epoch": 0.02748091603053435,
      "eval_loss": 2.113194227218628,
      "eval_runtime": 19.745,
      "eval_samples_per_second": 6.989,
      "eval_steps_per_second": 3.495,
      "step": 9
    },
    {
      "epoch": 0.030534351145038167,
      "grad_norm": 1.1229645013809204,
      "learning_rate": 0.0002,
      "loss": 1.9255,
      "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": 2052393617326080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}