trainer_state.json

{
  "best_metric": 0.6626126766204834,
  "best_model_checkpoint": "./results/roberta-base/checkpoint-8097",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 16194,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.618255853652954,
      "learning_rate": 4.672437244961049e-05,
      "loss": 0.3464,
      "step": 2699
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7697120458314488,
      "eval_conf_mat": [
        [
          6318,
          522
        ],
        [
          2533,
          3893
        ]
      ],
      "eval_f1": 0.7181994280970391,
      "eval_loss": 0.7109997272491455,
      "eval_precision": 0.8817667044167611,
      "eval_recall": 0.6058201058201058,
      "eval_runtime": 31.5248,
      "eval_samples_per_second": 420.812,
      "eval_steps_per_second": 13.164,
      "step": 2699
    },
    {
      "epoch": 2.0,
      "grad_norm": 30.145366668701172,
      "learning_rate": 4.338691727463831e-05,
      "loss": 0.1889,
      "step": 5398
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8014473089099955,
      "eval_conf_mat": [
        [
          6207,
          633
        ],
        [
          2001,
          4425
        ]
      ],
      "eval_f1": 0.7706374085684431,
      "eval_loss": 0.8169035911560059,
      "eval_precision": 0.8748517200474496,
      "eval_recall": 0.688608776844071,
      "eval_runtime": 31.2094,
      "eval_samples_per_second": 425.065,
      "eval_steps_per_second": 13.297,
      "step": 5398
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.833324432373047,
      "learning_rate": 4.004946209966614e-05,
      "loss": 0.1456,
      "step": 8097
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.818407960199005,
      "eval_conf_mat": [
        [
          6207,
          633
        ],
        [
          1776,
          4650
        ]
      ],
      "eval_f1": 0.7942608250064054,
      "eval_loss": 0.6626126766204834,
      "eval_precision": 0.8801817149346962,
      "eval_recall": 0.7236227824463118,
      "eval_runtime": 31.2622,
      "eval_samples_per_second": 424.346,
      "eval_steps_per_second": 13.275,
      "step": 8097
    },
    {
      "epoch": 4.0,
      "grad_norm": 16.268821716308594,
      "learning_rate": 3.6712006924693956e-05,
      "loss": 0.1181,
      "step": 10796
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7795115332428765,
      "eval_conf_mat": [
        [
          6320,
          520
        ],
        [
          2405,
          4021
        ]
      ],
      "eval_f1": 0.7332907814352148,
      "eval_loss": 1.1288779973983765,
      "eval_precision": 0.885487778022462,
      "eval_recall": 0.6257391845627139,
      "eval_runtime": 31.3467,
      "eval_samples_per_second": 423.202,
      "eval_steps_per_second": 13.239,
      "step": 10796
    },
    {
      "epoch": 5.0,
      "grad_norm": 152.2488555908203,
      "learning_rate": 3.337455174972178e-05,
      "loss": 0.0985,
      "step": 13495
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7281772953414745,
      "eval_conf_mat": [
        [
          6390,
          450
        ],
        [
          3156,
          3270
        ]
      ],
      "eval_f1": 0.644589000591366,
      "eval_loss": 1.4766658544540405,
      "eval_precision": 0.8790322580645161,
      "eval_recall": 0.5088702147525677,
      "eval_runtime": 31.2848,
      "eval_samples_per_second": 424.039,
      "eval_steps_per_second": 13.265,
      "step": 13495
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.3119850754737854,
      "learning_rate": 3.0037096574749602e-05,
      "loss": 0.0771,
      "step": 16194
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7473239861299563,
      "eval_conf_mat": [
        [
          6387,
          453
        ],
        [
          2899,
          3527
        ]
      ],
      "eval_f1": 0.6778781472227561,
      "eval_loss": 1.5410542488098145,
      "eval_precision": 0.8861809045226131,
      "eval_recall": 0.5488639900404606,
      "eval_runtime": 31.3637,
      "eval_samples_per_second": 422.973,
      "eval_steps_per_second": 13.232,
      "step": 16194
    }
  ],
  "logging_steps": 500,
  "max_steps": 40485,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.229063080284522e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}