pre-train-llama / config.json
Panhapich's picture
Add 86M llama-style checkpoint (epoch 1, step 71k), config, model code, and model card
8561bc9 verified
Raw
History Blame Contribute Delete
265 Bytes
{
"architecture": "llama_style_decoder_only_custom",
"model_config": {
"num_layers": 8,
"num_heads": 8,
"num_kv_heads": 4,
"hidden_dim": 768,
"max_seq_len": 512,
"vocab_size": 10000,
"dropout": 0.1
},
"num_parameters": 86235664
}