checkpoint-60/trainer_state.json

{
  "best_global_step": 60,
  "best_metric": 0.0092080757021904,
  "best_model_checkpoint": "/teamspace/studios/this_studio/DATN/output/medgemma_finetuned/checkpoint-60",
  "epoch": 4.288288288288288,
  "eval_steps": 15,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7207207207207207,
      "grad_norm": 0.6498702168464661,
      "learning_rate": 0.00018,
      "loss": 0.2477,
      "step": 10
    },
    {
      "epoch": 1.072072072072072,
      "eval_loss": 0.021378077566623688,
      "eval_runtime": 66.1391,
      "eval_samples_per_second": 0.242,
      "eval_steps_per_second": 0.121,
      "step": 15
    },
    {
      "epoch": 1.4324324324324325,
      "grad_norm": 2.6276116371154785,
      "learning_rate": 0.00019488259168373197,
      "loss": 0.0165,
      "step": 20
    },
    {
      "epoch": 2.144144144144144,
      "grad_norm": 0.3735731244087219,
      "learning_rate": 0.00017786415380497553,
      "loss": 0.0418,
      "step": 30
    },
    {
      "epoch": 2.144144144144144,
      "eval_loss": 0.019194122403860092,
      "eval_runtime": 23.7384,
      "eval_samples_per_second": 0.674,
      "eval_steps_per_second": 0.337,
      "step": 30
    },
    {
      "epoch": 2.864864864864865,
      "grad_norm": 0.07709665596485138,
      "learning_rate": 0.00015102700330609,
      "loss": 0.01,
      "step": 40
    },
    {
      "epoch": 3.2162162162162162,
      "eval_loss": 0.009275071322917938,
      "eval_runtime": 23.8923,
      "eval_samples_per_second": 0.67,
      "eval_steps_per_second": 0.335,
      "step": 45
    },
    {
      "epoch": 3.5765765765765765,
      "grad_norm": 0.09296473115682602,
      "learning_rate": 0.00011775531962543036,
      "loss": 0.0089,
      "step": 50
    },
    {
      "epoch": 4.288288288288288,
      "grad_norm": 0.06089961528778076,
      "learning_rate": 8.224468037456969e-05,
      "loss": 0.0088,
      "step": 60
    },
    {
      "epoch": 4.288288288288288,
      "eval_loss": 0.0092080757021904,
      "eval_runtime": 23.6439,
      "eval_samples_per_second": 0.677,
      "eval_steps_per_second": 0.338,
      "step": 60
    }
  ],
  "logging_steps": 10,
  "max_steps": 98,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 15,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 45,
        "early_stopping_threshold": 0.0001
      },
      "attributes": {
        "early_stopping_patience_counter": 1
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.559054638070912e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}