trainer_state.json

{
  "best_global_step": 10,
  "best_metric": 0.6849475502967834,
  "best_model_checkpoint": "distilbert_rand_50_v2_rte/checkpoint-10",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.684054434299469,
      "learning_rate": 4.9e-05,
      "loss": 0.6948,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5667870036101083,
      "eval_loss": 0.6849475502967834,
      "eval_runtime": 0.1259,
      "eval_samples_per_second": 2200.367,
      "eval_steps_per_second": 15.887,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8120187520980835,
      "learning_rate": 4.8e-05,
      "loss": 0.6795,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4981949458483754,
      "eval_loss": 0.716401994228363,
      "eval_runtime": 0.1296,
      "eval_samples_per_second": 2136.759,
      "eval_steps_per_second": 15.428,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.232509136199951,
      "learning_rate": 4.7e-05,
      "loss": 0.6306,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5234657039711191,
      "eval_loss": 0.7656673192977905,
      "eval_runtime": 0.1318,
      "eval_samples_per_second": 2101.495,
      "eval_steps_per_second": 15.173,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.697450876235962,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5263,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.900287926197052,
      "eval_runtime": 0.1369,
      "eval_samples_per_second": 2022.91,
      "eval_steps_per_second": 14.606,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.802547454833984,
      "learning_rate": 4.5e-05,
      "loss": 0.3928,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5126353790613718,
      "eval_loss": 1.1815361976623535,
      "eval_runtime": 0.1379,
      "eval_samples_per_second": 2008.836,
      "eval_steps_per_second": 14.504,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.6708805561065674,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.2431,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5306859205776173,
      "eval_loss": 1.4581865072250366,
      "eval_runtime": 0.1278,
      "eval_samples_per_second": 2166.673,
      "eval_steps_per_second": 15.644,
      "step": 60
    },
    {
      "epoch": 6.0,
      "step": 60,
      "total_flos": 989531467960320.0,
      "train_loss": 0.5278571565945943,
      "train_runtime": 22.7164,
      "train_samples_per_second": 5480.623,
      "train_steps_per_second": 22.011
    }
  ],
  "logging_steps": 1,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 989531467960320.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}