last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.07657945118059988,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025526483726866626,
      "eval_loss": 1.3599847555160522,
      "eval_runtime": 40.8475,
      "eval_samples_per_second": 4.039,
      "eval_steps_per_second": 2.032,
      "step": 1
    },
    {
      "epoch": 0.007657945118059987,
      "grad_norm": 269.10662841796875,
      "learning_rate": 4e-05,
      "loss": 5.519,
      "step": 3
    },
    {
      "epoch": 0.012763241863433313,
      "eval_loss": 1.356780767440796,
      "eval_runtime": 41.3585,
      "eval_samples_per_second": 3.99,
      "eval_steps_per_second": 2.007,
      "step": 5
    },
    {
      "epoch": 0.015315890236119975,
      "grad_norm": 1724.533935546875,
      "learning_rate": 8e-05,
      "loss": 5.8678,
      "step": 6
    },
    {
      "epoch": 0.02297383535417996,
      "grad_norm": 718.69921875,
      "learning_rate": 0.00012,
      "loss": 5.2231,
      "step": 9
    },
    {
      "epoch": 0.025526483726866625,
      "eval_loss": 1.3422503471374512,
      "eval_runtime": 41.4338,
      "eval_samples_per_second": 3.982,
      "eval_steps_per_second": 2.003,
      "step": 10
    },
    {
      "epoch": 0.03063178047223995,
      "grad_norm": 2000.920166015625,
      "learning_rate": 0.00016,
      "loss": 5.4501,
      "step": 12
    },
    {
      "epoch": 0.03828972559029994,
      "grad_norm": 143316.34375,
      "learning_rate": 0.0002,
      "loss": 5.3755,
      "step": 15
    },
    {
      "epoch": 0.03828972559029994,
      "eval_loss": 1.3301504850387573,
      "eval_runtime": 41.4574,
      "eval_samples_per_second": 3.98,
      "eval_steps_per_second": 2.002,
      "step": 15
    },
    {
      "epoch": 0.04594767070835992,
      "grad_norm": 2331.68115234375,
      "learning_rate": 0.00018090169943749476,
      "loss": 5.2684,
      "step": 18
    },
    {
      "epoch": 0.05105296745373325,
      "eval_loss": 1.322815179824829,
      "eval_runtime": 41.4853,
      "eval_samples_per_second": 3.977,
      "eval_steps_per_second": 2.001,
      "step": 20
    },
    {
      "epoch": 0.053605615826419914,
      "grad_norm": 1110465.0,
      "learning_rate": 0.00013090169943749476,
      "loss": 5.5578,
      "step": 21
    },
    {
      "epoch": 0.0612635609444799,
      "grad_norm": 72944.203125,
      "learning_rate": 6.909830056250527e-05,
      "loss": 5.022,
      "step": 24
    },
    {
      "epoch": 0.06381620931716656,
      "eval_loss": 1.3195682764053345,
      "eval_runtime": 41.4276,
      "eval_samples_per_second": 3.983,
      "eval_steps_per_second": 2.003,
      "step": 25
    },
    {
      "epoch": 0.06892150606253988,
      "grad_norm": 3819.193603515625,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 5.3914,
      "step": 27
    },
    {
      "epoch": 0.07657945118059988,
      "grad_norm": 4386.46533203125,
      "learning_rate": 0.0,
      "loss": 5.5024,
      "step": 30
    },
    {
      "epoch": 0.07657945118059988,
      "eval_loss": 1.3190549612045288,
      "eval_runtime": 41.4399,
      "eval_samples_per_second": 3.982,
      "eval_steps_per_second": 2.003,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.55315935232e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}