last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0026481297583581596,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000264812975835816,
      "grad_norm": 1.0843034982681274,
      "learning_rate": 2e-05,
      "loss": 1.2207,
      "step": 1
    },
    {
      "epoch": 0.000264812975835816,
      "eval_loss": 1.5418349504470825,
      "eval_runtime": 39.7224,
      "eval_samples_per_second": 40.028,
      "eval_steps_per_second": 20.014,
      "step": 1
    },
    {
      "epoch": 0.000529625951671632,
      "grad_norm": 1.033374547958374,
      "learning_rate": 4e-05,
      "loss": 1.3501,
      "step": 2
    },
    {
      "epoch": 0.0007944389275074479,
      "grad_norm": 1.3133691549301147,
      "learning_rate": 6e-05,
      "loss": 1.4165,
      "step": 3
    },
    {
      "epoch": 0.0007944389275074479,
      "eval_loss": 1.5384587049484253,
      "eval_runtime": 39.8393,
      "eval_samples_per_second": 39.91,
      "eval_steps_per_second": 19.955,
      "step": 3
    },
    {
      "epoch": 0.001059251903343264,
      "grad_norm": 1.2603070735931396,
      "learning_rate": 8e-05,
      "loss": 1.6498,
      "step": 4
    },
    {
      "epoch": 0.0013240648791790798,
      "grad_norm": 1.1261543035507202,
      "learning_rate": 0.0001,
      "loss": 1.3465,
      "step": 5
    },
    {
      "epoch": 0.0015888778550148957,
      "grad_norm": 1.2346397638320923,
      "learning_rate": 0.00012,
      "loss": 1.2569,
      "step": 6
    },
    {
      "epoch": 0.0015888778550148957,
      "eval_loss": 1.4900404214859009,
      "eval_runtime": 40.1043,
      "eval_samples_per_second": 39.647,
      "eval_steps_per_second": 19.823,
      "step": 6
    },
    {
      "epoch": 0.0018536908308507116,
      "grad_norm": 1.3048666715621948,
      "learning_rate": 0.00014,
      "loss": 1.5808,
      "step": 7
    },
    {
      "epoch": 0.002118503806686528,
      "grad_norm": 0.8697746992111206,
      "learning_rate": 0.00016,
      "loss": 1.2136,
      "step": 8
    },
    {
      "epoch": 0.0023833167825223437,
      "grad_norm": 0.9561617374420166,
      "learning_rate": 0.00018,
      "loss": 1.248,
      "step": 9
    },
    {
      "epoch": 0.0023833167825223437,
      "eval_loss": 1.3692142963409424,
      "eval_runtime": 40.223,
      "eval_samples_per_second": 39.53,
      "eval_steps_per_second": 19.765,
      "step": 9
    },
    {
      "epoch": 0.0026481297583581596,
      "grad_norm": 0.8304316401481628,
      "learning_rate": 0.0002,
      "loss": 0.9792,
      "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": 893296972922880.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}