LlamaGen / serve /fake_json /GPT-XXL.json
ShoufaChen's picture
remember json
e61c6be
raw
history blame
653 Bytes
{
"_name_or_path": "facebook/opt-125m",
"activation_dropout": 0.0,
"activation_function": "relu",
"architectures": [
"OPTForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 2,
"do_layer_norm_before": true,
"dropout": 0.1,
"eos_token_id": 2,
"ffn_dim": 3072,
"hidden_size": 1536,
"init_std": 0.02,
"layerdrop": 0.0,
"max_position_embeddings": 2048,
"model_type": "opt",
"num_attention_heads": 24,
"num_hidden_layers": 48,
"pad_token_id": 1,
"prefix": "</s>",
"torch_dtype": "bfloat16",
"transformers_version": "4.21.0.dev0",
"use_cache": true,
"vocab_size": 16384,
"word_embed_proj_dim": 768
}