{
  "norm_type": "Pre-LayerNorm",
  "vocab_size": 8192,
  "max_seq_len": 128,
  "d_model": 176,
  "n_heads": 8,
  "ffn_dim": 704,
  "num_blocks": 8,
  "epochs": 5,
  "train_losses": [
    2.7801686714166847,
    2.222144907208902,
    2.069231568489224,
    1.9887459733063775,
    1.9367085697778228
  ],
  "val_losses": [
    2.371724769221994,
    2.1539471743307708,
    2.060667110974099,
    2.008517362957909,
    1.9709118760985769
  ],
  "train_top1": [
    0.43638668426173,
    0.501614362949496,
    0.5221109322154497,
    0.5334409678465053,
    0.5410586588701694
  ],
  "train_top5": [
    0.6825656910253973,
    0.7609837558461504,
    0.7833337738495424,
    0.7951137859557944,
    0.802671929917193
  ],
  "val_top1": [
    0.4821287964004499,
    0.5104976060684722,
    0.5232406045398171,
    0.5306650370626749,
    0.5368395056387182
  ],
  "val_top5": [
    0.7390574976204898,
    0.771331948891004,
    0.7851424821897263,
    0.7932223616278734,
    0.7987990683856826
  ],
  "test_loss": 1.9709118760985769,
  "test_top1": 0.5368395056387182,
  "test_top5": 0.7987990683856826,
  "total_params": 5444480
}