last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0015617679212868967,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00015617679212868969,
      "grad_norm": 10.754196166992188,
      "learning_rate": 2e-05,
      "loss": 4.7032,
      "step": 1
    },
    {
      "epoch": 0.00015617679212868969,
      "eval_loss": 4.621094226837158,
      "eval_runtime": 128.0885,
      "eval_samples_per_second": 21.056,
      "eval_steps_per_second": 10.532,
      "step": 1
    },
    {
      "epoch": 0.00031235358425737937,
      "grad_norm": 11.617454528808594,
      "learning_rate": 4e-05,
      "loss": 4.5104,
      "step": 2
    },
    {
      "epoch": 0.00046853037638606906,
      "grad_norm": 10.591163635253906,
      "learning_rate": 6e-05,
      "loss": 4.8068,
      "step": 3
    },
    {
      "epoch": 0.00046853037638606906,
      "eval_loss": 4.157466411590576,
      "eval_runtime": 128.0132,
      "eval_samples_per_second": 21.068,
      "eval_steps_per_second": 10.538,
      "step": 3
    },
    {
      "epoch": 0.0006247071685147587,
      "grad_norm": 10.794215202331543,
      "learning_rate": 8e-05,
      "loss": 4.1871,
      "step": 4
    },
    {
      "epoch": 0.0007808839606434484,
      "grad_norm": 11.163933753967285,
      "learning_rate": 0.0001,
      "loss": 2.9558,
      "step": 5
    },
    {
      "epoch": 0.0009370607527721381,
      "grad_norm": 8.859397888183594,
      "learning_rate": 0.00012,
      "loss": 1.8565,
      "step": 6
    },
    {
      "epoch": 0.0009370607527721381,
      "eval_loss": 1.340795636177063,
      "eval_runtime": 128.0023,
      "eval_samples_per_second": 21.07,
      "eval_steps_per_second": 10.539,
      "step": 6
    },
    {
      "epoch": 0.0010932375449008277,
      "grad_norm": 4.494658470153809,
      "learning_rate": 0.00014,
      "loss": 1.4069,
      "step": 7
    },
    {
      "epoch": 0.0012494143370295175,
      "grad_norm": 5.664268493652344,
      "learning_rate": 0.00016,
      "loss": 1.1381,
      "step": 8
    },
    {
      "epoch": 0.001405591129158207,
      "grad_norm": 4.099843502044678,
      "learning_rate": 0.00018,
      "loss": 0.5486,
      "step": 9
    },
    {
      "epoch": 0.001405591129158207,
      "eval_loss": 0.2859042286872864,
      "eval_runtime": 128.2715,
      "eval_samples_per_second": 21.026,
      "eval_steps_per_second": 10.517,
      "step": 9
    },
    {
      "epoch": 0.0015617679212868967,
      "grad_norm": 9.554787635803223,
      "learning_rate": 0.0002,
      "loss": 0.4099,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2052393617326080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}