last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 7,
  "global_step": 27,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.037037037037037035,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.037037037037037035,
      "eval_loss": NaN,
      "eval_runtime": 18.7818,
      "eval_samples_per_second": 9.69,
      "eval_steps_per_second": 2.449,
      "step": 1
    },
    {
      "epoch": 0.07407407407407407,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": NaN,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": NaN,
      "learning_rate": 9.842915805643155e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.18518518518518517,
      "grad_norm": NaN,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": NaN,
      "learning_rate": 9.381533400219318e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.25925925925925924,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.25925925925925924,
      "eval_loss": NaN,
      "eval_runtime": 18.0856,
      "eval_samples_per_second": 10.063,
      "eval_steps_per_second": 2.543,
      "step": 7
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": NaN,
      "learning_rate": 8.644843137107059e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": NaN,
      "learning_rate": 8.18711994874345e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": NaN,
      "learning_rate": 7.679133974894983e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.4074074074074074,
      "grad_norm": NaN,
      "learning_rate": 7.128896457825364e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.48148148148148145,
      "grad_norm": NaN,
      "learning_rate": 5.9369065729286245e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.5185185185185185,
      "grad_norm": NaN,
      "learning_rate": 5.313952597646568e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.5185185185185185,
      "eval_loss": NaN,
      "eval_runtime": 18.089,
      "eval_samples_per_second": 10.061,
      "eval_steps_per_second": 2.543,
      "step": 14
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": NaN,
      "learning_rate": 4.6860474023534335e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": NaN,
      "learning_rate": 4.063093427071376e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.6296296296296297,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": NaN,
      "learning_rate": 2.8711035421746367e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.7037037037037037,
      "grad_norm": NaN,
      "learning_rate": 2.3208660251050158e-05,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": NaN,
      "learning_rate": 1.8128800512565513e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.7777777777777778,
      "grad_norm": NaN,
      "learning_rate": 1.3551568628929434e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.7777777777777778,
      "eval_loss": NaN,
      "eval_runtime": 18.0827,
      "eval_samples_per_second": 10.065,
      "eval_steps_per_second": 2.544,
      "step": 21
    },
    {
      "epoch": 0.8148148148148148,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.8518518518518519,
      "grad_norm": NaN,
      "learning_rate": 6.184665997806832e-06,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": NaN,
      "learning_rate": 3.511175705587433e-06,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": NaN,
      "learning_rate": 1.5708419435684462e-06,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.9629629629629629,
      "grad_norm": NaN,
      "learning_rate": 3.9426493427611177e-07,
      "loss": 0.0,
      "step": 26
    },
    {
      "epoch": 1.0,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 27
    }
  ],
  "logging_steps": 1,
  "max_steps": 27,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 7,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.03969416538882e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}