enigma-1.5b / enigma /config_enigma.json
shivendrra's picture
added training and model files
f44acc0 verified
raw
history blame
222 Bytes
{
"batch_size": 10,
"block_size": 512,
"max_iters": 5000,
"eval_interval": 50,
"learning_rate": 3e-5,
"eval_iters": 100,
"d_model": 384,
"n_head": 12,
"n_layer": 12,
"dropout": 0.2,
"norm_eps": 1e-5
}