last-checkpoint/trainer_state.json

{
  "best_metric": NaN,
  "best_model_checkpoint": "miner_id_24/checkpoint-50",
  "epoch": 0.08090614886731391,
  "eval_steps": 50,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005393743257820927,
      "eval_loss": NaN,
      "eval_runtime": 14.1111,
      "eval_samples_per_second": 55.347,
      "eval_steps_per_second": 13.89,
      "step": 1
    },
    {
      "epoch": 0.005393743257820928,
      "grad_norm": NaN,
      "learning_rate": 5.05e-06,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.010787486515641856,
      "grad_norm": NaN,
      "learning_rate": 1.01e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.016181229773462782,
      "grad_norm": NaN,
      "learning_rate": 9.323076923076924e-06,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.021574973031283712,
      "grad_norm": NaN,
      "learning_rate": 8.546153846153847e-06,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.02696871628910464,
      "grad_norm": NaN,
      "learning_rate": 7.76923076923077e-06,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.02696871628910464,
      "eval_loss": NaN,
      "eval_runtime": 13.1829,
      "eval_samples_per_second": 59.243,
      "eval_steps_per_second": 14.868,
      "step": 50
    },
    {
      "epoch": 0.032362459546925564,
      "grad_norm": NaN,
      "learning_rate": 6.992307692307692e-06,
      "loss": 0.0,
      "step": 60
    },
    {
      "epoch": 0.037756202804746494,
      "grad_norm": NaN,
      "learning_rate": 6.215384615384615e-06,
      "loss": 0.0,
      "step": 70
    },
    {
      "epoch": 0.043149946062567425,
      "grad_norm": NaN,
      "learning_rate": 5.438461538461538e-06,
      "loss": 0.0,
      "step": 80
    },
    {
      "epoch": 0.04854368932038835,
      "grad_norm": NaN,
      "learning_rate": 4.661538461538462e-06,
      "loss": 0.0,
      "step": 90
    },
    {
      "epoch": 0.05393743257820928,
      "grad_norm": NaN,
      "learning_rate": 3.884615384615385e-06,
      "loss": 0.0,
      "step": 100
    },
    {
      "epoch": 0.05393743257820928,
      "eval_loss": NaN,
      "eval_runtime": 13.1871,
      "eval_samples_per_second": 59.225,
      "eval_steps_per_second": 14.863,
      "step": 100
    },
    {
      "epoch": 0.0593311758360302,
      "grad_norm": NaN,
      "learning_rate": 3.1076923076923076e-06,
      "loss": 0.0,
      "step": 110
    },
    {
      "epoch": 0.06472491909385113,
      "grad_norm": NaN,
      "learning_rate": 2.330769230769231e-06,
      "loss": 0.0,
      "step": 120
    },
    {
      "epoch": 0.07011866235167206,
      "grad_norm": NaN,
      "learning_rate": 1.5538461538461538e-06,
      "loss": 0.0,
      "step": 130
    },
    {
      "epoch": 0.07551240560949299,
      "grad_norm": NaN,
      "learning_rate": 7.769230769230769e-07,
      "loss": 0.0,
      "step": 140
    },
    {
      "epoch": 0.08090614886731391,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 150
    },
    {
      "epoch": 0.08090614886731391,
      "eval_loss": NaN,
      "eval_runtime": 13.194,
      "eval_samples_per_second": 59.194,
      "eval_steps_per_second": 14.855,
      "step": 150
    }
  ],
  "logging_steps": 10,
  "max_steps": 150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 2
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7557290404085760.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}