checkpoint-100/trainer_state.json

{
  "best_global_step": 100,
  "best_metric": 0.23627299070358276,
  "best_model_checkpoint": "./phi3-payments-reverse-finetuned\\checkpoint-100",
  "epoch": 2.0,
  "eval_steps": 50,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 1.149180293083191,
      "learning_rate": 3.6e-05,
      "loss": 2.0358,
      "step": 10
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.7529087662696838,
      "learning_rate": 7.6e-05,
      "loss": 1.798,
      "step": 20
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.9188901782035828,
      "learning_rate": 0.000116,
      "loss": 1.2619,
      "step": 30
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5950552225112915,
      "learning_rate": 0.00015600000000000002,
      "loss": 0.7113,
      "step": 40
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5578396320343018,
      "learning_rate": 0.000196,
      "loss": 0.4846,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.40434688329696655,
      "eval_runtime": 23.0633,
      "eval_samples_per_second": 2.168,
      "eval_steps_per_second": 2.168,
      "step": 50
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.7984590530395508,
      "learning_rate": 0.000182,
      "loss": 0.3761,
      "step": 60
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.4209927022457123,
      "learning_rate": 0.000162,
      "loss": 0.2922,
      "step": 70
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.361447811126709,
      "learning_rate": 0.000142,
      "loss": 0.2698,
      "step": 80
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.3857899010181427,
      "learning_rate": 0.000122,
      "loss": 0.2565,
      "step": 90
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.30109161138534546,
      "learning_rate": 0.00010200000000000001,
      "loss": 0.2459,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.23627299070358276,
      "eval_runtime": 20.8349,
      "eval_samples_per_second": 2.4,
      "eval_steps_per_second": 2.4,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9170514345984000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}