last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.090909090909091,
  "eval_steps": 50,
  "global_step": 17,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 0.7348418831825256,
      "learning_rate": 1e-05,
      "loss": 1.9656,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "eval_loss": 1.9155492782592773,
      "eval_runtime": 0.8849,
      "eval_samples_per_second": 11.3,
      "eval_steps_per_second": 3.39,
      "step": 1
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.729081392288208,
      "learning_rate": 2e-05,
      "loss": 1.9391,
      "step": 2
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 0.7109789252281189,
      "learning_rate": 3e-05,
      "loss": 1.8572,
      "step": 3
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.6369128823280334,
      "learning_rate": 4e-05,
      "loss": 1.9209,
      "step": 4
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.5599340796470642,
      "learning_rate": 5e-05,
      "loss": 1.8946,
      "step": 5
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.894423246383667,
      "learning_rate": 6e-05,
      "loss": 2.9971,
      "step": 6
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 0.48314210772514343,
      "learning_rate": 7e-05,
      "loss": 1.7987,
      "step": 7
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.5811020731925964,
      "learning_rate": 8e-05,
      "loss": 1.8732,
      "step": 8
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 0.6997883319854736,
      "learning_rate": 9e-05,
      "loss": 1.6008,
      "step": 9
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.5665810108184814,
      "learning_rate": 0.0001,
      "loss": 1.6587,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7635406255722046,
      "learning_rate": 9.504844339512095e-05,
      "loss": 2.2619,
      "step": 11
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.47633466124534607,
      "learning_rate": 8.117449009293668e-05,
      "loss": 1.5668,
      "step": 12
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.4734385311603546,
      "learning_rate": 6.112604669781572e-05,
      "loss": 1.4911,
      "step": 13
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.4208812713623047,
      "learning_rate": 3.887395330218429e-05,
      "loss": 1.4513,
      "step": 14
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.40034040808677673,
      "learning_rate": 1.8825509907063327e-05,
      "loss": 1.4712,
      "step": 15
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.42380931973457336,
      "learning_rate": 4.951556604879048e-06,
      "loss": 1.4123,
      "step": 16
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.6289786696434021,
      "learning_rate": 0.0,
      "loss": 2.4548,
      "step": 17
    }
  ],
  "logging_steps": 1,
  "max_steps": 17,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "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": 2.290019746185216e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}