checkpoint-12000/trainer_state.json

{
  "best_metric": 3.5764389038085938,
  "best_model_checkpoint": "/Users/frapadovani/Desktop/babyLM_controlled/models_trained/convergence_french/random_sentence_french/checkpoint-12000",
  "epoch": 0.35531341604240074,
  "eval_steps": 2000,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.059218902673733455,
      "grad_norm": 1.0812722444534302,
      "learning_rate": 0.0001,
      "loss": 4.7523,
      "step": 2000
    },
    {
      "epoch": 0.059218902673733455,
      "eval_loss": 4.0236735343933105,
      "eval_runtime": 5.173,
      "eval_samples_per_second": 425.477,
      "eval_steps_per_second": 26.677,
      "step": 2000
    },
    {
      "epoch": 0.11843780534746691,
      "grad_norm": 1.2557106018066406,
      "learning_rate": 0.0001,
      "loss": 3.8958,
      "step": 4000
    },
    {
      "epoch": 0.11843780534746691,
      "eval_loss": 3.8087611198425293,
      "eval_runtime": 5.2666,
      "eval_samples_per_second": 417.913,
      "eval_steps_per_second": 26.203,
      "step": 4000
    },
    {
      "epoch": 0.17765670802120037,
      "grad_norm": 1.3326870203018188,
      "learning_rate": 0.0001,
      "loss": 3.7344,
      "step": 6000
    },
    {
      "epoch": 0.17765670802120037,
      "eval_loss": 3.704986333847046,
      "eval_runtime": 5.2277,
      "eval_samples_per_second": 421.023,
      "eval_steps_per_second": 26.398,
      "step": 6000
    },
    {
      "epoch": 0.23687561069493382,
      "grad_norm": 1.38993239402771,
      "learning_rate": 0.0001,
      "loss": 3.6379,
      "step": 8000
    },
    {
      "epoch": 0.23687561069493382,
      "eval_loss": 3.6455252170562744,
      "eval_runtime": 5.1485,
      "eval_samples_per_second": 427.5,
      "eval_steps_per_second": 26.804,
      "step": 8000
    },
    {
      "epoch": 0.29609451336866727,
      "grad_norm": 1.3884963989257812,
      "learning_rate": 0.0001,
      "loss": 3.5687,
      "step": 10000
    },
    {
      "epoch": 0.29609451336866727,
      "eval_loss": 3.6062421798706055,
      "eval_runtime": 5.1466,
      "eval_samples_per_second": 427.661,
      "eval_steps_per_second": 26.814,
      "step": 10000
    },
    {
      "epoch": 0.35531341604240074,
      "grad_norm": 1.435062289237976,
      "learning_rate": 0.0001,
      "loss": 3.5165,
      "step": 12000
    },
    {
      "epoch": 0.35531341604240074,
      "eval_loss": 3.5764389038085938,
      "eval_runtime": 5.132,
      "eval_samples_per_second": 428.881,
      "eval_steps_per_second": 26.89,
      "step": 12000
    }
  ],
  "logging_steps": 2000,
  "max_steps": 33773,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2000,
  "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": 233928916992000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}