trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4433673918247223,
  "best_model_checkpoint": "distilbert_km_100_v2_sst2/checkpoint-264",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1584,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 8.28586196899414,
      "learning_rate": 4.9e-05,
      "loss": 0.4334,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8096330275229358,
      "eval_loss": 0.4433673918247223,
      "eval_runtime": 0.3703,
      "eval_samples_per_second": 2354.927,
      "eval_steps_per_second": 10.802,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.146331787109375,
      "learning_rate": 4.8e-05,
      "loss": 0.2294,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7970183486238532,
      "eval_loss": 0.5247591733932495,
      "eval_runtime": 0.3657,
      "eval_samples_per_second": 2384.201,
      "eval_steps_per_second": 10.937,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.0153398513793945,
      "learning_rate": 4.7e-05,
      "loss": 0.1671,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7901376146788991,
      "eval_loss": 0.5308615565299988,
      "eval_runtime": 0.4193,
      "eval_samples_per_second": 2079.662,
      "eval_steps_per_second": 9.54,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.469273567199707,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1254,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8004587155963303,
      "eval_loss": 0.6569430828094482,
      "eval_runtime": 0.3625,
      "eval_samples_per_second": 2405.774,
      "eval_steps_per_second": 11.036,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 7.875831127166748,
      "learning_rate": 4.5e-05,
      "loss": 0.095,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8096330275229358,
      "eval_loss": 0.6505075693130493,
      "eval_runtime": 0.374,
      "eval_samples_per_second": 2331.634,
      "eval_steps_per_second": 10.696,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.4298948049545288,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0726,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7935779816513762,
      "eval_loss": 0.7870606780052185,
      "eval_runtime": 0.369,
      "eval_samples_per_second": 2362.837,
      "eval_steps_per_second": 10.839,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.18715316599065607,
      "train_runtime": 378.7397,
      "train_samples_per_second": 8891.199,
      "train_steps_per_second": 34.852
    }
  ],
  "logging_steps": 1,
  "max_steps": 13200,
  "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": 2.676464049624883e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}