trainer_state.json

{
  "best_global_step": 21000,
  "best_metric": 0.0023173103109002113,
  "best_model_checkpoint": "./t5gemma-finetuned_full_dataset_small\\checkpoint-21000",
  "epoch": 0.40385634733569,
  "eval_steps": 7000,
  "global_step": 21000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03846250927006571,
      "grad_norm": 5.125,
      "learning_rate": 1.992715874723852e-05,
      "loss": 0.049,
      "step": 2000
    },
    {
      "epoch": 0.07692501854013142,
      "grad_norm": 1.0078125,
      "learning_rate": 1.97095516230889e-05,
      "loss": 0.0128,
      "step": 4000
    },
    {
      "epoch": 0.11538752781019714,
      "grad_norm": 3.0,
      "learning_rate": 1.9350351933080317e-05,
      "loss": 0.0082,
      "step": 6000
    },
    {
      "epoch": 0.13461878244522998,
      "eval_loss": 0.00584625406190753,
      "eval_runtime": 9078.4693,
      "eval_samples_per_second": 22.911,
      "eval_steps_per_second": 22.911,
      "step": 7000
    },
    {
      "epoch": 0.15385003708026285,
      "grad_norm": 0.59375,
      "learning_rate": 1.885479781876022e-05,
      "loss": 0.0061,
      "step": 8000
    },
    {
      "epoch": 0.19231254635032854,
      "grad_norm": 1.125,
      "learning_rate": 1.823011585261026e-05,
      "loss": 0.0051,
      "step": 10000
    },
    {
      "epoch": 0.23077505562039427,
      "grad_norm": 0.369140625,
      "learning_rate": 1.748541565429842e-05,
      "loss": 0.0044,
      "step": 12000
    },
    {
      "epoch": 0.26923756489045997,
      "grad_norm": 1.1171875,
      "learning_rate": 1.6631557046808075e-05,
      "loss": 0.0039,
      "step": 14000
    },
    {
      "epoch": 0.26923756489045997,
      "eval_loss": 0.003004432423040271,
      "eval_runtime": 9508.738,
      "eval_samples_per_second": 21.874,
      "eval_steps_per_second": 21.874,
      "step": 14000
    },
    {
      "epoch": 0.3077000741605257,
      "grad_norm": 0.25390625,
      "learning_rate": 1.568099168968113e-05,
      "loss": 0.0037,
      "step": 16000
    },
    {
      "epoch": 0.3461625834305914,
      "grad_norm": 0.1611328125,
      "learning_rate": 1.464758149880794e-05,
      "loss": 0.0034,
      "step": 18000
    },
    {
      "epoch": 0.3846250927006571,
      "grad_norm": 1.125,
      "learning_rate": 1.3546396500713988e-05,
      "loss": 0.0032,
      "step": 20000
    },
    {
      "epoch": 0.40385634733569,
      "eval_loss": 0.0023173103109002113,
      "eval_runtime": 9393.4852,
      "eval_samples_per_second": 22.142,
      "eval_steps_per_second": 22.142,
      "step": 21000
    }
  ],
  "logging_steps": 2000,
  "max_steps": 51999,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 7000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.393692126347264e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}