XVERSE-13B-LLAMA / config.json
genggui001's picture
First version of the your-model-name model and tokenizer.
a93d500
raw
history blame contribute delete
503 Bytes
{
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 2,
"eos_token_id": 3,
"hidden_act": "silu",
"hidden_size": 5120,
"initializer_range": 0.02,
"intermediate_size": 13824,
"max_position_embeddings": 8192,
"model_type": "llama",
"num_attention_heads": 40,
"num_hidden_layers": 40,
"pad_token_id": 1,
"rms_norm_eps": 1e-06,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.28.1",
"use_cache": true,
"vocab_size": 100352
}