trainer_state.json

{
  "best_global_step": 80,
  "best_metric": 0.10693139582872391,
  "best_model_checkpoint": "/content/models/gemma_jigsaw_instruction/checkpoint-80",
  "epoch": 1.5714285714285714,
  "eval_steps": 20,
  "global_step": 80,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.0030554473400115,
      "epoch": 0.39408866995073893,
      "grad_norm": 2.5103867053985596,
      "learning_rate": 8.758169934640524e-06,
      "loss": 0.3751,
      "mean_token_accuracy": 0.859375,
      "num_tokens": 72532.0,
      "step": 20
    },
    {
      "epoch": 0.39408866995073893,
      "eval_entropy": 2.080273380646339,
      "eval_loss": 0.14906036853790283,
      "eval_mean_token_accuracy": 0.9230769230769231,
      "eval_num_tokens": 72532.0,
      "eval_runtime": 4.4741,
      "eval_samples_per_second": 45.372,
      "eval_steps_per_second": 5.811,
      "step": 20
    },
    {
      "entropy": 2.026494912803173,
      "epoch": 0.7881773399014779,
      "grad_norm": 8.653420448303223,
      "learning_rate": 7.450980392156863e-06,
      "loss": 0.1611,
      "mean_token_accuracy": 0.916015625,
      "num_tokens": 145048.0,
      "step": 40
    },
    {
      "epoch": 0.7881773399014779,
      "eval_entropy": 2.0305226903695326,
      "eval_loss": 0.16201180219650269,
      "eval_mean_token_accuracy": 0.9254807692307693,
      "eval_num_tokens": 145048.0,
      "eval_runtime": 4.4358,
      "eval_samples_per_second": 45.764,
      "eval_steps_per_second": 5.861,
      "step": 40
    },
    {
      "entropy": 2.0197064559670945,
      "epoch": 1.1773399014778325,
      "grad_norm": 2.4835453033447266,
      "learning_rate": 6.143790849673204e-06,
      "loss": 0.1224,
      "mean_token_accuracy": 0.9418512658227848,
      "num_tokens": 216334.0,
      "step": 60
    },
    {
      "epoch": 1.1773399014778325,
      "eval_entropy": 2.0450107271854696,
      "eval_loss": 0.1152966096997261,
      "eval_mean_token_accuracy": 0.9407051274409661,
      "eval_num_tokens": 216334.0,
      "eval_runtime": 4.4391,
      "eval_samples_per_second": 45.73,
      "eval_steps_per_second": 5.857,
      "step": 60
    },
    {
      "entropy": 1.9920476481318474,
      "epoch": 1.5714285714285714,
      "grad_norm": 5.939174175262451,
      "learning_rate": 4.836601307189543e-06,
      "loss": 0.0992,
      "mean_token_accuracy": 0.95625,
      "num_tokens": 289742.0,
      "step": 80
    },
    {
      "epoch": 1.5714285714285714,
      "eval_entropy": 2.0168408017892103,
      "eval_loss": 0.10693139582872391,
      "eval_mean_token_accuracy": 0.9499198725590339,
      "eval_num_tokens": 289742.0,
      "eval_runtime": 4.4376,
      "eval_samples_per_second": 45.746,
      "eval_steps_per_second": 5.859,
      "step": 80
    }
  ],
  "logging_steps": 20,
  "max_steps": 153,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5508744073159680.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}