27 lines
678 B
JSON
27 lines
678 B
JSON
{
|
|
"architectures": [
|
|
"AquilaForCausalLM"
|
|
],
|
|
"auto_map": {
|
|
"AutoConfig": "configuration_aquila.AquilaConfig",
|
|
"AutoModelForCausalLM": "modeling_aquila.AquilaForCausalLM"
|
|
},
|
|
"bos_token_id": 100006,
|
|
"eos_token_id": 100007,
|
|
"hidden_act": "silu",
|
|
"hidden_size": 4096,
|
|
"initializer_range": 0.02,
|
|
"intermediate_size": 11008,
|
|
"max_position_embeddings": 2048,
|
|
"model_type": "aquila",
|
|
"num_attention_heads": 32,
|
|
"num_hidden_layers": 32,
|
|
"pad_token_id": 0,
|
|
"rms_norm_eps": 1e-05,
|
|
"rope_scaling": null,
|
|
"tie_word_embeddings": false,
|
|
"torch_dtype": "float16",
|
|
"transformers_version": "4.31.0",
|
|
"use_cache": true,
|
|
"vocab_size": 100008
|
|
} |