trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 179,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11173184357541899,
      "grad_norm": 2.0458016395568848,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 2.0107,
      "step": 20
    },
    {
      "epoch": 0.22346368715083798,
      "grad_norm": 2.3176522254943848,
      "learning_rate": 4.9903533134293035e-06,
      "loss": 1.9625,
      "step": 40
    },
    {
      "epoch": 0.33519553072625696,
      "grad_norm": 2.3477518558502197,
      "learning_rate": 4.660472094042121e-06,
      "loss": 1.7282,
      "step": 60
    },
    {
      "epoch": 0.44692737430167595,
      "grad_norm": 2.8214402198791504,
      "learning_rate": 3.92016186682789e-06,
      "loss": 1.4047,
      "step": 80
    },
    {
      "epoch": 0.5586592178770949,
      "grad_norm": 1.2627031803131104,
      "learning_rate": 2.9100607788275547e-06,
      "loss": 1.1131,
      "step": 100
    },
    {
      "epoch": 0.6703910614525139,
      "grad_norm": 1.03634512424469,
      "learning_rate": 1.8220596619089576e-06,
      "loss": 0.9726,
      "step": 120
    },
    {
      "epoch": 0.7821229050279329,
      "grad_norm": 0.9175835847854614,
      "learning_rate": 8.628481651367876e-07,
      "loss": 0.8899,
      "step": 140
    },
    {
      "epoch": 0.8938547486033519,
      "grad_norm": 0.9105198979377747,
      "learning_rate": 2.1464952759020857e-07,
      "loss": 0.8588,
      "step": 160
    },
    {
      "epoch": 1.0,
      "step": 179,
      "total_flos": 1.6486718999887872e+16,
      "train_loss": 1.3161976137640756,
      "train_runtime": 493.8113,
      "train_samples_per_second": 1.45,
      "train_steps_per_second": 0.362
    }
  ],
  "logging_steps": 20,
  "max_steps": 179,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 1.6486718999887872e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}