tinystories_updated / config.json
Jumpr's picture
Push model using huggingface_hub.
714a69f verified
Raw
History Blame Contribute Delete
182 Bytes
{
"batch_size": 12,
"block_num": 16,
"embed_dims": 1024,
"head_size": 64,
"iterations": 40000,
"lr": 0.0003,
"num_heads": 16,
"seq_len": 2048,
"vocab_size": 49152
}