trainer_state.json

{
  "best_metric": 0.7935978659553184,
  "best_model_checkpoint": "logs/ecthr_b/roberta-base/seed_1/checkpoint-1410",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2256,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.19746574759483337,
      "eval_macro-f1": 0.6393379993931348,
      "eval_micro-f1": 0.7408184679958028,
      "eval_runtime": 9.7693,
      "eval_samples_per_second": 102.361,
      "eval_steps_per_second": 3.276,
      "step": 282
    },
    {
      "epoch": 1.773049645390071,
      "grad_norm": 1.1872084140777588,
      "learning_rate": 2.7340425531914897e-05,
      "loss": 0.1811,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.19539476931095123,
      "eval_macro-f1": 0.6540544618364194,
      "eval_micro-f1": 0.7559311017224569,
      "eval_runtime": 9.1514,
      "eval_samples_per_second": 109.272,
      "eval_steps_per_second": 3.497,
      "step": 564
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.178602933883667,
      "eval_macro-f1": 0.7063419905042477,
      "eval_micro-f1": 0.7832625040863027,
      "eval_runtime": 8.9637,
      "eval_samples_per_second": 111.562,
      "eval_steps_per_second": 3.57,
      "step": 846
    },
    {
      "epoch": 3.546099290780142,
      "grad_norm": 1.4385334253311157,
      "learning_rate": 2.4686170212765958e-05,
      "loss": 0.1167,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.1745893508195877,
      "eval_macro-f1": 0.7303617301778567,
      "eval_micro-f1": 0.7927686642115835,
      "eval_runtime": 9.0428,
      "eval_samples_per_second": 110.585,
      "eval_steps_per_second": 3.539,
      "step": 1128
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.1818280965089798,
      "eval_macro-f1": 0.7269570388508974,
      "eval_micro-f1": 0.7935978659553184,
      "eval_runtime": 8.4498,
      "eval_samples_per_second": 118.346,
      "eval_steps_per_second": 3.787,
      "step": 1410
    },
    {
      "epoch": 5.319148936170213,
      "grad_norm": 1.5338271856307983,
      "learning_rate": 2.2026595744680854e-05,
      "loss": 0.0921,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.1933058500289917,
      "eval_macro-f1": 0.7235350370954329,
      "eval_micro-f1": 0.7809836065573771,
      "eval_runtime": 8.4822,
      "eval_samples_per_second": 117.894,
      "eval_steps_per_second": 3.773,
      "step": 1692
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.19006936252117157,
      "eval_macro-f1": 0.7326251101371736,
      "eval_micro-f1": 0.7851602023608769,
      "eval_runtime": 8.4696,
      "eval_samples_per_second": 118.069,
      "eval_steps_per_second": 3.778,
      "step": 1974
    },
    {
      "epoch": 7.092198581560283,
      "grad_norm": 2.809311866760254,
      "learning_rate": 1.9367021276595743e-05,
      "loss": 0.0721,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.19798380136489868,
      "eval_macro-f1": 0.7335571414521224,
      "eval_micro-f1": 0.7897877984084881,
      "eval_runtime": 8.5312,
      "eval_samples_per_second": 117.217,
      "eval_steps_per_second": 3.751,
      "step": 2256
    },
    {
      "epoch": 8.0,
      "step": 2256,
      "total_flos": 3.542404433675551e+17,
      "train_loss": 0.10894472971030161,
      "train_runtime": 1334.5785,
      "train_samples_per_second": 134.874,
      "train_steps_per_second": 4.226
    }
  ],
  "logging_steps": 500,
  "max_steps": 5640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 3.542404433675551e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}