trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.886075949367089,
  "eval_steps": 100,
  "global_step": 195,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25316455696202533,
      "grad_norm": 15.541438102722168,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 4.3307,
      "step": 10
    },
    {
      "epoch": 0.5063291139240507,
      "grad_norm": 16.98396110534668,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 4.3055,
      "step": 20
    },
    {
      "epoch": 0.759493670886076,
      "grad_norm": 17.21928596496582,
      "learning_rate": 3e-06,
      "loss": 4.1895,
      "step": 30
    },
    {
      "epoch": 1.0,
      "grad_norm": 13.936151504516602,
      "learning_rate": 4.000000000000001e-06,
      "loss": 3.8896,
      "step": 40
    },
    {
      "epoch": 1.2531645569620253,
      "grad_norm": 16.210145950317383,
      "learning_rate": 5e-06,
      "loss": 3.894,
      "step": 50
    },
    {
      "epoch": 1.5063291139240507,
      "grad_norm": 15.298253059387207,
      "learning_rate": 6e-06,
      "loss": 3.6968,
      "step": 60
    },
    {
      "epoch": 1.759493670886076,
      "grad_norm": 16.198633193969727,
      "learning_rate": 7.000000000000001e-06,
      "loss": 3.529,
      "step": 70
    },
    {
      "epoch": 2.0,
      "grad_norm": 12.727923393249512,
      "learning_rate": 8.000000000000001e-06,
      "loss": 3.1801,
      "step": 80
    },
    {
      "epoch": 2.2531645569620253,
      "grad_norm": 13.769573211669922,
      "learning_rate": 9e-06,
      "loss": 3.049,
      "step": 90
    },
    {
      "epoch": 2.5063291139240507,
      "grad_norm": 14.311426162719727,
      "learning_rate": 1e-05,
      "loss": 2.9882,
      "step": 100
    },
    {
      "epoch": 2.5063291139240507,
      "eval_accuracy": 0.22857142857142856,
      "eval_f1": 0.14422788951090837,
      "eval_loss": 2.718834161758423,
      "eval_runtime": 22.0561,
      "eval_samples_per_second": 3.174,
      "eval_steps_per_second": 0.408,
      "step": 100
    },
    {
      "epoch": 2.759493670886076,
      "grad_norm": 14.486056327819824,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 2.755,
      "step": 110
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.645160675048828,
      "learning_rate": 1.2e-05,
      "loss": 2.4722,
      "step": 120
    },
    {
      "epoch": 3.2531645569620253,
      "grad_norm": 14.1466064453125,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 2.4231,
      "step": 130
    },
    {
      "epoch": 3.5063291139240507,
      "grad_norm": 14.645957946777344,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 2.3087,
      "step": 140
    },
    {
      "epoch": 3.759493670886076,
      "grad_norm": 12.975454330444336,
      "learning_rate": 1.5e-05,
      "loss": 2.193,
      "step": 150
    },
    {
      "epoch": 4.0,
      "grad_norm": 11.665339469909668,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.0295,
      "step": 160
    },
    {
      "epoch": 4.253164556962025,
      "grad_norm": 12.780133247375488,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 1.8715,
      "step": 170
    },
    {
      "epoch": 4.506329113924051,
      "grad_norm": 15.635778427124023,
      "learning_rate": 1.8e-05,
      "loss": 1.8326,
      "step": 180
    },
    {
      "epoch": 4.759493670886076,
      "grad_norm": 12.904759407043457,
      "learning_rate": 1.9e-05,
      "loss": 1.7675,
      "step": 190
    }
  ],
  "logging_steps": 10,
  "max_steps": 195,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1673975515914240.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}