llama-lite-134m / config.json
skeskinen's picture
Upload LlamaForCausalLM
2cb9900
raw
history blame contribute delete
535 Bytes
{
"_name_or_path": "./output",
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 2048,
"max_position_embeddings": 2048,
"model_type": "llama",
"num_attention_heads": 6,
"num_hidden_layers": 12,
"pad_token_id": 0,
"rms_norm_eps": 1e-12,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.29.0.dev0",
"use_cache": true,
"vocab_size": 32000
}