trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6693227091633465,
  "eval_steps": 15,
  "global_step": 105,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01593625498007968,
      "grad_norm": 0.15623033046722412,
      "learning_rate": 0.0005,
      "loss": 0.7895,
      "step": 1
    },
    {
      "epoch": 0.23904382470119523,
      "grad_norm": 0.4118459224700928,
      "learning_rate": 0.0005,
      "loss": 0.639,
      "step": 15
    },
    {
      "epoch": 0.23904382470119523,
      "eval_loss": 0.5786148309707642,
      "eval_runtime": 203.1738,
      "eval_samples_per_second": 0.305,
      "eval_steps_per_second": 0.305,
      "step": 15
    },
    {
      "epoch": 0.47808764940239046,
      "grad_norm": 0.28019416332244873,
      "learning_rate": 0.0005,
      "loss": 0.6105,
      "step": 30
    },
    {
      "epoch": 0.47808764940239046,
      "eval_loss": 0.5625983476638794,
      "eval_runtime": 203.0773,
      "eval_samples_per_second": 0.305,
      "eval_steps_per_second": 0.305,
      "step": 30
    },
    {
      "epoch": 0.7171314741035857,
      "grad_norm": 0.302608460187912,
      "learning_rate": 0.0005,
      "loss": 0.5912,
      "step": 45
    },
    {
      "epoch": 0.7171314741035857,
      "eval_loss": 0.5549846887588501,
      "eval_runtime": 203.1082,
      "eval_samples_per_second": 0.305,
      "eval_steps_per_second": 0.305,
      "step": 45
    },
    {
      "epoch": 0.9561752988047809,
      "grad_norm": 0.27707505226135254,
      "learning_rate": 0.0005,
      "loss": 0.5778,
      "step": 60
    },
    {
      "epoch": 0.9561752988047809,
      "eval_loss": 0.5472670197486877,
      "eval_runtime": 203.1439,
      "eval_samples_per_second": 0.305,
      "eval_steps_per_second": 0.305,
      "step": 60
    },
    {
      "epoch": 1.1912350597609562,
      "grad_norm": 0.28894054889678955,
      "learning_rate": 0.0005,
      "loss": 0.4824,
      "step": 75
    },
    {
      "epoch": 1.1912350597609562,
      "eval_loss": 0.5558595061302185,
      "eval_runtime": 202.9842,
      "eval_samples_per_second": 0.305,
      "eval_steps_per_second": 0.305,
      "step": 75
    },
    {
      "epoch": 1.4302788844621515,
      "grad_norm": 0.31184110045433044,
      "learning_rate": 0.0005,
      "loss": 0.4642,
      "step": 90
    },
    {
      "epoch": 1.4302788844621515,
      "eval_loss": 0.5655969977378845,
      "eval_runtime": 203.0879,
      "eval_samples_per_second": 0.305,
      "eval_steps_per_second": 0.305,
      "step": 90
    },
    {
      "epoch": 1.6693227091633465,
      "grad_norm": 0.33053600788116455,
      "learning_rate": 0.0005,
      "loss": 0.4727,
      "step": 105
    },
    {
      "epoch": 1.6693227091633465,
      "eval_loss": 0.5597427487373352,
      "eval_runtime": 203.0614,
      "eval_samples_per_second": 0.305,
      "eval_steps_per_second": 0.305,
      "step": 105
    }
  ],
  "logging_steps": 15,
  "max_steps": 248,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9661979026785075e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}