last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0004216198211910338,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.432396423820677e-06,
      "eval_loss": NaN,
      "eval_runtime": 2042.9263,
      "eval_samples_per_second": 24.442,
      "eval_steps_per_second": 12.221,
      "step": 1
    },
    {
      "epoch": 4.216198211910338e-05,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 8.432396423820676e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 8.432396423820676e-05,
      "eval_loss": NaN,
      "eval_runtime": 2055.2843,
      "eval_samples_per_second": 24.295,
      "eval_steps_per_second": 12.148,
      "step": 10
    },
    {
      "epoch": 0.00012648594635731016,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.00016864792847641352,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.00016864792847641352,
      "eval_loss": NaN,
      "eval_runtime": 2067.6727,
      "eval_samples_per_second": 24.149,
      "eval_steps_per_second": 12.075,
      "step": 20
    },
    {
      "epoch": 0.0002108099105955169,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.0002529718927146203,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0002529718927146203,
      "eval_loss": NaN,
      "eval_runtime": 2046.6025,
      "eval_samples_per_second": 24.398,
      "eval_steps_per_second": 12.199,
      "step": 30
    },
    {
      "epoch": 0.0002951338748337237,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.00033729585695282705,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.00033729585695282705,
      "eval_loss": NaN,
      "eval_runtime": 2060.3373,
      "eval_samples_per_second": 24.235,
      "eval_steps_per_second": 12.118,
      "step": 40
    },
    {
      "epoch": 0.00037945783907193046,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.0004216198211910338,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.0004216198211910338,
      "eval_loss": NaN,
      "eval_runtime": 2058.08,
      "eval_samples_per_second": 24.262,
      "eval_steps_per_second": 12.131,
      "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": 2302517635448832.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}