last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00014804095553034747,
  "eval_steps": 5,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.402047776517374e-06,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 7.402047776517374e-06,
      "eval_loss": NaN,
      "eval_runtime": 393.5324,
      "eval_samples_per_second": 72.274,
      "eval_steps_per_second": 72.274,
      "step": 1
    },
    {
      "epoch": 1.4804095553034748e-05,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 2.2206143329552122e-05,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 2.9608191106069495e-05,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 3.701023888258687e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 3.701023888258687e-05,
      "eval_loss": NaN,
      "eval_runtime": 392.1863,
      "eval_samples_per_second": 72.522,
      "eval_steps_per_second": 72.522,
      "step": 5
    },
    {
      "epoch": 4.4412286659104244e-05,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 5.1814334435621614e-05,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 5.921638221213899e-05,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 6.661842998865637e-05,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 7.402047776517374e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 7.402047776517374e-05,
      "eval_loss": NaN,
      "eval_runtime": 392.3695,
      "eval_samples_per_second": 72.488,
      "eval_steps_per_second": 72.488,
      "step": 10
    },
    {
      "epoch": 8.14225255416911e-05,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 8.882457331820849e-05,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 9.622662109472586e-05,
      "grad_norm": NaN,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.00010362866887124323,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.0001110307166477606,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0001110307166477606,
      "eval_loss": NaN,
      "eval_runtime": 270.2917,
      "eval_samples_per_second": 105.227,
      "eval_steps_per_second": 105.227,
      "step": 15
    },
    {
      "epoch": 0.00011843276442427798,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.00012583481220079535,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.00013323685997731273,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0001406389077538301,
      "grad_norm": NaN,
      "learning_rate": 4.8943483704846475e-06,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.00014804095553034747,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.00014804095553034747,
      "eval_loss": NaN,
      "eval_runtime": 168.5642,
      "eval_samples_per_second": 168.731,
      "eval_steps_per_second": 168.731,
      "step": 20
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5357017497600.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}