Sheared-LLaMA-2.7B / config.json
xiamengzhou
update
c38d2e6
raw
history blame
624 Bytes
{
"_name_or_path": "meta-llama/Llama-2-7b-hf",
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2560,
"initializer_range": 0.02,
"intermediate_size": 6912,
"max_position_embeddings": 4096,
"model_type": "llama",
"num_attention_heads": 20,
"num_hidden_layers": 32,
"num_key_value_heads": 20,
"pad_token_id": 0,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.28.1",
"use_cache": true,
"vocab_size": 32000
}