last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004860857941426662,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001620285980475554,
      "eval_loss": NaN,
      "eval_runtime": 396.7996,
      "eval_samples_per_second": 6.55,
      "eval_steps_per_second": 3.276,
      "step": 1
    },
    {
      "epoch": 0.00048608579414266616,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.000810142990237777,
      "eval_loss": NaN,
      "eval_runtime": 396.7818,
      "eval_samples_per_second": 6.55,
      "eval_steps_per_second": 3.276,
      "step": 5
    },
    {
      "epoch": 0.0009721715882853323,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0014582573824279985,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.001620285980475554,
      "eval_loss": NaN,
      "eval_runtime": 396.7973,
      "eval_samples_per_second": 6.55,
      "eval_steps_per_second": 3.276,
      "step": 10
    },
    {
      "epoch": 0.0019443431765706646,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.002430428970713331,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.002430428970713331,
      "eval_loss": NaN,
      "eval_runtime": 396.8677,
      "eval_samples_per_second": 6.549,
      "eval_steps_per_second": 3.276,
      "step": 15
    },
    {
      "epoch": 0.002916514764855997,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.003240571960951108,
      "eval_loss": NaN,
      "eval_runtime": 396.8551,
      "eval_samples_per_second": 6.549,
      "eval_steps_per_second": 3.276,
      "step": 20
    },
    {
      "epoch": 0.0034026005589986633,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0038886863531413293,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.004050714951188885,
      "eval_loss": NaN,
      "eval_runtime": 396.7977,
      "eval_samples_per_second": 6.55,
      "eval_steps_per_second": 3.276,
      "step": 25
    },
    {
      "epoch": 0.004374772147283996,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.004860857941426662,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.004860857941426662,
      "eval_loss": NaN,
      "eval_runtime": 396.7824,
      "eval_samples_per_second": 6.55,
      "eval_steps_per_second": 3.276,
      "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": 2.110703430480691e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}