27 lines
583 B
JSON
27 lines
583 B
JSON
|
{
|
||
|
"activation_dropout": 0.0,
|
||
|
"activation_function": "relu",
|
||
|
"architectures": [
|
||
|
"OPTModel"
|
||
|
],
|
||
|
"attention_dropout": 0.0,
|
||
|
"bos_token_id": 0,
|
||
|
"d_model": 2048,
|
||
|
"do_layer_norm_before": true,
|
||
|
"dropout": 0.1,
|
||
|
"eos_token_id": 2,
|
||
|
"ffn_dim": 8192,
|
||
|
"init_std": 0.02,
|
||
|
"layerdrop": 0.0,
|
||
|
"max_position_embeddings": 2048,
|
||
|
"model_type": "opt",
|
||
|
"num_attention_heads": 32,
|
||
|
"num_hidden_layers": 24,
|
||
|
"pad_token_id": 1,
|
||
|
"torch_dtype": "float16",
|
||
|
"transformers_version": "4.19.0.dev0",
|
||
|
"use_cache": true,
|
||
|
"vocab_size": 50272,
|
||
|
"word_embed_proj_dim": 2048
|
||
|
}
|