last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02566295979469632,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001710863986313088,
      "grad_norm": 1.584230661392212,
      "learning_rate": 1e-05,
      "loss": 0.7921,
      "step": 1
    },
    {
      "epoch": 0.001710863986313088,
      "eval_loss": 0.9855340719223022,
      "eval_runtime": 68.2666,
      "eval_samples_per_second": 14.429,
      "eval_steps_per_second": 7.222,
      "step": 1
    },
    {
      "epoch": 0.003421727972626176,
      "grad_norm": 1.6465091705322266,
      "learning_rate": 2e-05,
      "loss": 0.8332,
      "step": 2
    },
    {
      "epoch": 0.0051325919589392645,
      "grad_norm": 1.3933875560760498,
      "learning_rate": 3e-05,
      "loss": 0.8667,
      "step": 3
    },
    {
      "epoch": 0.006843455945252352,
      "grad_norm": 1.1915502548217773,
      "learning_rate": 4e-05,
      "loss": 0.7275,
      "step": 4
    },
    {
      "epoch": 0.00855431993156544,
      "grad_norm": 1.950568437576294,
      "learning_rate": 5e-05,
      "loss": 0.8782,
      "step": 5
    },
    {
      "epoch": 0.00855431993156544,
      "eval_loss": 0.9573764204978943,
      "eval_runtime": 4.8483,
      "eval_samples_per_second": 203.163,
      "eval_steps_per_second": 101.685,
      "step": 5
    },
    {
      "epoch": 0.010265183917878529,
      "grad_norm": 1.549292802810669,
      "learning_rate": 6e-05,
      "loss": 0.7648,
      "step": 6
    },
    {
      "epoch": 0.011976047904191617,
      "grad_norm": 1.3967912197113037,
      "learning_rate": 7e-05,
      "loss": 0.7261,
      "step": 7
    },
    {
      "epoch": 0.013686911890504704,
      "grad_norm": 3.4337575435638428,
      "learning_rate": 8e-05,
      "loss": 0.7624,
      "step": 8
    },
    {
      "epoch": 0.015397775876817793,
      "grad_norm": 1.231019139289856,
      "learning_rate": 9e-05,
      "loss": 0.678,
      "step": 9
    },
    {
      "epoch": 0.01710863986313088,
      "grad_norm": 1.2229334115982056,
      "learning_rate": 0.0001,
      "loss": 0.6854,
      "step": 10
    },
    {
      "epoch": 0.01710863986313088,
      "eval_loss": 0.8504759669303894,
      "eval_runtime": 4.9,
      "eval_samples_per_second": 201.021,
      "eval_steps_per_second": 100.613,
      "step": 10
    },
    {
      "epoch": 0.018819503849443968,
      "grad_norm": 1.2083159685134888,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.7023,
      "step": 11
    },
    {
      "epoch": 0.020530367835757058,
      "grad_norm": 1.0155587196350098,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.6187,
      "step": 12
    },
    {
      "epoch": 0.022241231822070145,
      "grad_norm": 1.0483033657073975,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.7,
      "step": 13
    },
    {
      "epoch": 0.023952095808383235,
      "grad_norm": 0.8496240973472595,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.5521,
      "step": 14
    },
    {
      "epoch": 0.02566295979469632,
      "grad_norm": 1.384743571281433,
      "learning_rate": 5e-05,
      "loss": 0.7186,
      "step": 15
    },
    {
      "epoch": 0.02566295979469632,
      "eval_loss": 0.7239178419113159,
      "eval_runtime": 4.9498,
      "eval_samples_per_second": 198.999,
      "eval_steps_per_second": 99.6,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 304347272970240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}