checkpoint-12000/trainer_state.json

{
  "best_metric": 3.1521711349487305,
  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_recent/mlm_unmasking/fr_mlm_new/childes_mlm_unmasking_context_42/checkpoint-12000",
  "epoch": 21.0896309314587,
  "eval_steps": 2000,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.51493848857645,
      "eval_loss": 5.561293601989746,
      "eval_runtime": 1.7076,
      "eval_samples_per_second": 1305.334,
      "eval_steps_per_second": 81.986,
      "step": 2000
    },
    {
      "epoch": 7.0298769771529,
      "grad_norm": 1.3643518686294556,
      "learning_rate": 4e-05,
      "loss": 6.2537,
      "step": 4000
    },
    {
      "epoch": 7.0298769771529,
      "eval_loss": 5.49988317489624,
      "eval_runtime": 1.7023,
      "eval_samples_per_second": 1309.415,
      "eval_steps_per_second": 82.242,
      "step": 4000
    },
    {
      "epoch": 10.54481546572935,
      "eval_loss": 5.408681392669678,
      "eval_runtime": 1.7195,
      "eval_samples_per_second": 1296.321,
      "eval_steps_per_second": 81.42,
      "step": 6000
    },
    {
      "epoch": 14.0597539543058,
      "grad_norm": 3.336256265640259,
      "learning_rate": 8e-05,
      "loss": 5.329,
      "step": 8000
    },
    {
      "epoch": 14.0597539543058,
      "eval_loss": 4.505647659301758,
      "eval_runtime": 1.6952,
      "eval_samples_per_second": 1314.863,
      "eval_steps_per_second": 82.584,
      "step": 8000
    },
    {
      "epoch": 17.57469244288225,
      "eval_loss": 3.6001980304718018,
      "eval_runtime": 1.6952,
      "eval_samples_per_second": 1314.862,
      "eval_steps_per_second": 82.584,
      "step": 10000
    },
    {
      "epoch": 21.0896309314587,
      "grad_norm": 2.967787504196167,
      "learning_rate": 0.00012,
      "loss": 3.7571,
      "step": 12000
    },
    {
      "epoch": 21.0896309314587,
      "eval_loss": 3.1521711349487305,
      "eval_runtime": 1.6965,
      "eval_samples_per_second": 1313.906,
      "eval_steps_per_second": 82.524,
      "step": 12000
    }
  ],
  "logging_steps": 4000,
  "max_steps": 400000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 703,
  "save_steps": 4000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.001
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2502641283563520.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}