quality_chunk/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.995249406175772,
  "eval_steps": 500,
  "global_step": 420,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24,
      "global_step": 50,
      "learning_rate": 0.0001,
      "loss": 0.5683,
      "step": 50
    },
    {
      "epoch": 0.48,
      "global_step": 100,
      "learning_rate": 0.0001,
      "loss": 0.5393,
      "step": 100
    },
    {
      "epoch": 0.71,
      "global_step": 150,
      "learning_rate": 0.0001,
      "loss": 0.5376,
      "step": 150
    },
    {
      "epoch": 0.95,
      "global_step": 200,
      "learning_rate": 0.0001,
      "loss": 0.5331,
      "step": 200
    },
    {
      "epoch": 1.19,
      "global_step": 250,
      "learning_rate": 0.0001,
      "loss": 0.3011,
      "step": 250
    },
    {
      "epoch": 1.43,
      "global_step": 300,
      "learning_rate": 0.0001,
      "loss": 0.1997,
      "step": 300
    },
    {
      "epoch": 1.66,
      "global_step": 350,
      "learning_rate": 0.0001,
      "loss": 0.1919,
      "step": 350
    },
    {
      "epoch": 1.9,
      "global_step": 400,
      "learning_rate": 0.0001,
      "loss": 0.1817,
      "step": 400
    },
    {
      "epoch": 2.0,
      "step": 420,
      "total_flos": 2.272886224060416e+17,
      "train_loss": 0.37379822503952753,
      "train_runtime": 12484.1415,
      "train_samples_per_second": 0.404,
      "train_steps_per_second": 0.034
    }
  ],
  "logging_steps": 50,
  "max_steps": 420,
  "num_train_epochs": 2,
  "save_steps": 200,
  "total_flos": 2.272886224060416e+17,
  "trial_name": null,
  "trial_params": null
}