last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0008209135947395856,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.6418271894791713e-05,
      "eval_loss": NaN,
      "eval_runtime": 1483.5964,
      "eval_samples_per_second": 17.286,
      "eval_steps_per_second": 8.643,
      "step": 1
    },
    {
      "epoch": 8.209135947395857e-05,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00016418271894791714,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.00016418271894791714,
      "eval_loss": NaN,
      "eval_runtime": 1481.348,
      "eval_samples_per_second": 17.313,
      "eval_steps_per_second": 8.656,
      "step": 10
    },
    {
      "epoch": 0.0002462740784218757,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0003283654378958343,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0003283654378958343,
      "eval_loss": NaN,
      "eval_runtime": 1481.0789,
      "eval_samples_per_second": 17.316,
      "eval_steps_per_second": 8.658,
      "step": 20
    },
    {
      "epoch": 0.0004104567973697928,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.0004925481568437514,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0004925481568437514,
      "eval_loss": NaN,
      "eval_runtime": 1481.9318,
      "eval_samples_per_second": 17.306,
      "eval_steps_per_second": 8.653,
      "step": 30
    },
    {
      "epoch": 0.00057463951631771,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.0006567308757916686,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.0006567308757916686,
      "eval_loss": NaN,
      "eval_runtime": 1484.0218,
      "eval_samples_per_second": 17.281,
      "eval_steps_per_second": 8.641,
      "step": 40
    },
    {
      "epoch": 0.0007388222352656272,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.0008209135947395856,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.0008209135947395856,
      "eval_loss": NaN,
      "eval_runtime": 1487.701,
      "eval_samples_per_second": 17.239,
      "eval_steps_per_second": 8.619,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8713180230451200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}