last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.19011406844106463,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0076045627376425855,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0076045627376425855,
      "eval_loss": NaN,
      "eval_runtime": 12.9046,
      "eval_samples_per_second": 8.602,
      "eval_steps_per_second": 1.085,
      "step": 1
    },
    {
      "epoch": 0.015209125475285171,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.022813688212927757,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.030418250950570342,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.030418250950570342,
      "eval_loss": NaN,
      "eval_runtime": 12.1151,
      "eval_samples_per_second": 9.162,
      "eval_steps_per_second": 1.156,
      "step": 4
    },
    {
      "epoch": 0.03802281368821293,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.045627376425855515,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.053231939163498096,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.060836501901140684,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.060836501901140684,
      "eval_loss": NaN,
      "eval_runtime": 12.1124,
      "eval_samples_per_second": 9.164,
      "eval_steps_per_second": 1.156,
      "step": 8
    },
    {
      "epoch": 0.06844106463878327,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.07604562737642585,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.08365019011406843,
      "grad_norm": NaN,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.09125475285171103,
      "grad_norm": NaN,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.09125475285171103,
      "eval_loss": NaN,
      "eval_runtime": 12.113,
      "eval_samples_per_second": 9.164,
      "eval_steps_per_second": 1.156,
      "step": 12
    },
    {
      "epoch": 0.09885931558935361,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.10646387832699619,
      "grad_norm": NaN,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.11406844106463879,
      "grad_norm": NaN,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.12167300380228137,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.12167300380228137,
      "eval_loss": NaN,
      "eval_runtime": 12.1136,
      "eval_samples_per_second": 9.163,
      "eval_steps_per_second": 1.156,
      "step": 16
    },
    {
      "epoch": 0.12927756653992395,
      "grad_norm": NaN,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.13688212927756654,
      "grad_norm": NaN,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.1444866920152091,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.1520912547528517,
      "grad_norm": NaN,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.1520912547528517,
      "eval_loss": NaN,
      "eval_runtime": 12.1127,
      "eval_samples_per_second": 9.164,
      "eval_steps_per_second": 1.156,
      "step": 20
    },
    {
      "epoch": 0.1596958174904943,
      "grad_norm": NaN,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.16730038022813687,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.17490494296577946,
      "grad_norm": NaN,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.18250950570342206,
      "grad_norm": NaN,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.18250950570342206,
      "eval_loss": NaN,
      "eval_runtime": 12.1113,
      "eval_samples_per_second": 9.165,
      "eval_steps_per_second": 1.156,
      "step": 24
    },
    {
      "epoch": 0.19011406844106463,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.85471820890112e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}