last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004300381658872225,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002866921105914817,
      "grad_norm": 2.3434934616088867,
      "learning_rate": 1e-05,
      "loss": 2.3633,
      "step": 1
    },
    {
      "epoch": 0.0002866921105914817,
      "eval_loss": 2.5438878536224365,
      "eval_runtime": 193.7172,
      "eval_samples_per_second": 30.328,
      "eval_steps_per_second": 15.166,
      "step": 1
    },
    {
      "epoch": 0.0005733842211829634,
      "grad_norm": 1.747134804725647,
      "learning_rate": 2e-05,
      "loss": 2.4938,
      "step": 2
    },
    {
      "epoch": 0.000860076331774445,
      "grad_norm": 1.7026760578155518,
      "learning_rate": 3e-05,
      "loss": 2.338,
      "step": 3
    },
    {
      "epoch": 0.0011467684423659267,
      "grad_norm": 1.9191794395446777,
      "learning_rate": 4e-05,
      "loss": 2.6796,
      "step": 4
    },
    {
      "epoch": 0.0014334605529574083,
      "grad_norm": 1.7047806978225708,
      "learning_rate": 5e-05,
      "loss": 2.4081,
      "step": 5
    },
    {
      "epoch": 0.0014334605529574083,
      "eval_loss": 2.463175058364868,
      "eval_runtime": 48.5493,
      "eval_samples_per_second": 121.011,
      "eval_steps_per_second": 60.516,
      "step": 5
    },
    {
      "epoch": 0.00172015266354889,
      "grad_norm": 1.7617462873458862,
      "learning_rate": 6e-05,
      "loss": 2.2562,
      "step": 6
    },
    {
      "epoch": 0.0020068447741403714,
      "grad_norm": 1.7860451936721802,
      "learning_rate": 7e-05,
      "loss": 2.4667,
      "step": 7
    },
    {
      "epoch": 0.0022935368847318534,
      "grad_norm": 1.786962628364563,
      "learning_rate": 8e-05,
      "loss": 2.3607,
      "step": 8
    },
    {
      "epoch": 0.002580228995323335,
      "grad_norm": 1.9283440113067627,
      "learning_rate": 9e-05,
      "loss": 2.2896,
      "step": 9
    },
    {
      "epoch": 0.0028669211059148166,
      "grad_norm": 1.9662737846374512,
      "learning_rate": 0.0001,
      "loss": 2.1083,
      "step": 10
    },
    {
      "epoch": 0.0028669211059148166,
      "eval_loss": 2.109614610671997,
      "eval_runtime": 48.5494,
      "eval_samples_per_second": 121.011,
      "eval_steps_per_second": 60.516,
      "step": 10
    },
    {
      "epoch": 0.003153613216506298,
      "grad_norm": 1.9880238771438599,
      "learning_rate": 9.755282581475769e-05,
      "loss": 2.2276,
      "step": 11
    },
    {
      "epoch": 0.00344030532709778,
      "grad_norm": 1.6718554496765137,
      "learning_rate": 9.045084971874738e-05,
      "loss": 2.0454,
      "step": 12
    },
    {
      "epoch": 0.0037269974376892617,
      "grad_norm": 1.6015005111694336,
      "learning_rate": 7.938926261462366e-05,
      "loss": 1.7847,
      "step": 13
    },
    {
      "epoch": 0.004013689548280743,
      "grad_norm": 1.6074224710464478,
      "learning_rate": 6.545084971874738e-05,
      "loss": 1.807,
      "step": 14
    },
    {
      "epoch": 0.004300381658872225,
      "grad_norm": 1.5804719924926758,
      "learning_rate": 5e-05,
      "loss": 1.5977,
      "step": 15
    },
    {
      "epoch": 0.004300381658872225,
      "eval_loss": 1.6562390327453613,
      "eval_runtime": 52.2266,
      "eval_samples_per_second": 112.491,
      "eval_steps_per_second": 56.255,
      "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": 1547541468413952.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}