last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0168067226890756,
  "eval_steps": 6,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03361344537815126,
      "eval_loss": 0.4333963394165039,
      "eval_runtime": 0.9506,
      "eval_samples_per_second": 13.675,
      "eval_steps_per_second": 7.364,
      "step": 1
    },
    {
      "epoch": 0.16806722689075632,
      "grad_norm": 0.773507297039032,
      "learning_rate": 5e-05,
      "loss": 0.4071,
      "step": 5
    },
    {
      "epoch": 0.20168067226890757,
      "eval_loss": 0.3601323664188385,
      "eval_runtime": 0.9296,
      "eval_samples_per_second": 13.984,
      "eval_steps_per_second": 7.53,
      "step": 6
    },
    {
      "epoch": 0.33613445378151263,
      "grad_norm": 0.6221334934234619,
      "learning_rate": 0.0001,
      "loss": 0.2676,
      "step": 10
    },
    {
      "epoch": 0.40336134453781514,
      "eval_loss": 0.019643442705273628,
      "eval_runtime": 0.9567,
      "eval_samples_per_second": 13.589,
      "eval_steps_per_second": 7.317,
      "step": 12
    },
    {
      "epoch": 0.5042016806722689,
      "grad_norm": 0.08144775032997131,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0289,
      "step": 15
    },
    {
      "epoch": 0.6050420168067226,
      "eval_loss": 0.001077459310181439,
      "eval_runtime": 1.0113,
      "eval_samples_per_second": 12.854,
      "eval_steps_per_second": 6.921,
      "step": 18
    },
    {
      "epoch": 0.6722689075630253,
      "grad_norm": 0.04066217690706253,
      "learning_rate": 5e-05,
      "loss": 0.0018,
      "step": 20
    },
    {
      "epoch": 0.8067226890756303,
      "eval_loss": 0.0004625740693882108,
      "eval_runtime": 0.9911,
      "eval_samples_per_second": 13.117,
      "eval_steps_per_second": 7.063,
      "step": 24
    },
    {
      "epoch": 0.8403361344537815,
      "grad_norm": 0.008725620806217194,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0006,
      "step": 25
    },
    {
      "epoch": 1.0168067226890756,
      "grad_norm": 0.014241045340895653,
      "learning_rate": 0.0,
      "loss": 0.0005,
      "step": 30
    },
    {
      "epoch": 1.0168067226890756,
      "eval_loss": 0.0004074503667652607,
      "eval_runtime": 0.9569,
      "eval_samples_per_second": 13.585,
      "eval_steps_per_second": 7.315,
      "step": 30
    }
  ],
  "logging_steps": 5,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4886202427637760.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}