last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0003571205371092878,
  "eval_steps": 4,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.7856026855464392e-05,
      "grad_norm": NaN,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 1.7856026855464392e-05,
      "eval_loss": NaN,
      "eval_runtime": 1669.0335,
      "eval_samples_per_second": 14.129,
      "eval_steps_per_second": 3.533,
      "step": 1
    },
    {
      "epoch": 3.5712053710928784e-05,
      "grad_norm": NaN,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 5.356808056639317e-05,
      "grad_norm": NaN,
      "learning_rate": 6e-06,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 7.142410742185757e-05,
      "grad_norm": NaN,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 7.142410742185757e-05,
      "eval_loss": NaN,
      "eval_runtime": 1669.3122,
      "eval_samples_per_second": 14.126,
      "eval_steps_per_second": 3.532,
      "step": 4
    },
    {
      "epoch": 8.928013427732195e-05,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00010713616113278634,
      "grad_norm": NaN,
      "learning_rate": 9.960573506572391e-06,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.00012499218798825074,
      "grad_norm": NaN,
      "learning_rate": 9.842915805643156e-06,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.00014284821484371514,
      "grad_norm": NaN,
      "learning_rate": 9.648882429441258e-06,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.00014284821484371514,
      "eval_loss": NaN,
      "eval_runtime": 1669.7018,
      "eval_samples_per_second": 14.123,
      "eval_steps_per_second": 3.531,
      "step": 8
    },
    {
      "epoch": 0.0001607042416991795,
      "grad_norm": NaN,
      "learning_rate": 9.381533400219319e-06,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0001785602685546439,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0001964162954101083,
      "grad_norm": NaN,
      "learning_rate": 8.644843137107058e-06,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.00021427232226557268,
      "grad_norm": NaN,
      "learning_rate": 8.18711994874345e-06,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.00021427232226557268,
      "eval_loss": NaN,
      "eval_runtime": 1669.3622,
      "eval_samples_per_second": 14.126,
      "eval_steps_per_second": 3.532,
      "step": 12
    },
    {
      "epoch": 0.00023212834912103708,
      "grad_norm": NaN,
      "learning_rate": 7.679133974894984e-06,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.0002499843759765015,
      "grad_norm": NaN,
      "learning_rate": 7.128896457825364e-06,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.00026784040283196585,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0002856964296874303,
      "grad_norm": NaN,
      "learning_rate": 5.936906572928625e-06,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.0002856964296874303,
      "eval_loss": NaN,
      "eval_runtime": 1667.6779,
      "eval_samples_per_second": 14.14,
      "eval_steps_per_second": 3.535,
      "step": 16
    },
    {
      "epoch": 0.00030355245654289465,
      "grad_norm": NaN,
      "learning_rate": 5.3139525976465675e-06,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.000321408483398359,
      "grad_norm": NaN,
      "learning_rate": 4.686047402353433e-06,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.00033926451025382344,
      "grad_norm": NaN,
      "learning_rate": 4.063093427071376e-06,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.0003571205371092878,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0003571205371092878,
      "eval_loss": NaN,
      "eval_runtime": 1667.572,
      "eval_samples_per_second": 14.141,
      "eval_steps_per_second": 3.536,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2285970464440320.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}