last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0024018253872943435,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.006084624314479e-05,
      "eval_loss": NaN,
      "eval_runtime": 792.3243,
      "eval_samples_per_second": 6.639,
      "eval_steps_per_second": 3.319,
      "step": 1
    },
    {
      "epoch": 0.00024018253872943438,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00040030423121572396,
      "eval_loss": NaN,
      "eval_runtime": 796.7994,
      "eval_samples_per_second": 6.601,
      "eval_steps_per_second": 3.301,
      "step": 5
    },
    {
      "epoch": 0.00048036507745886876,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0007205476161883031,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0008006084624314479,
      "eval_loss": NaN,
      "eval_runtime": 790.9522,
      "eval_samples_per_second": 6.65,
      "eval_steps_per_second": 3.325,
      "step": 10
    },
    {
      "epoch": 0.0009607301549177375,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0012009126936471718,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0012009126936471718,
      "eval_loss": NaN,
      "eval_runtime": 795.7453,
      "eval_samples_per_second": 6.61,
      "eval_steps_per_second": 3.305,
      "step": 15
    },
    {
      "epoch": 0.0014410952323766061,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0016012169248628958,
      "eval_loss": NaN,
      "eval_runtime": 791.2546,
      "eval_samples_per_second": 6.648,
      "eval_steps_per_second": 3.324,
      "step": 20
    },
    {
      "epoch": 0.0016812777711060407,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.001921460309835475,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.0020015211560786197,
      "eval_loss": NaN,
      "eval_runtime": 793.5872,
      "eval_samples_per_second": 6.628,
      "eval_steps_per_second": 3.314,
      "step": 25
    },
    {
      "epoch": 0.0021616428485649094,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.0024018253872943435,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0024018253872943435,
      "eval_loss": NaN,
      "eval_runtime": 794.3563,
      "eval_samples_per_second": 6.622,
      "eval_steps_per_second": 3.311,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9801873028546560.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}