smol_llama-101M-GQA / config.json
pszemraj's picture
Upload LlamaForCausalLM
4c964e8
raw
history blame contribute delete
670 Bytes
{
"_name_or_path": "BEE-spoke-data/smol_llama-GQA-KV8-v13_KIx2",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 1024,
"model_type": "llama",
"num_attention_heads": 24,
"num_hidden_layers": 6,
"num_key_value_heads": 8,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.34.1",
"use_cache": true,
"vocab_size": 32128
}