saiga2-13b-4bit / config.json
gurgutan's picture
AutoGPTQ model for IlyaGusev/saiga2_13b_lora: 4bits, gr128, desc_act=False
eef4dc4
raw
history blame
614 Bytes
{
"_name_or_path": "saiga2-merged",
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 5120,
"initializer_range": 0.02,
"intermediate_size": 13824,
"max_position_embeddings": 4096,
"model_type": "llama",
"num_attention_heads": 40,
"num_hidden_layers": 40,
"num_key_value_heads": 40,
"pad_token_id": 0,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.29.2",
"use_cache": true,
"vocab_size": 32000
}