初始化项目,由ModelHub XC社区提供模型
Model: CMSManhattan/JiRackUltra_1b Source: Original Platform
This commit is contained in:
21
config.json
Normal file
21
config.json
Normal file
@@ -0,0 +1,21 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen2ForCausalLM"
|
||||
],
|
||||
"model_type": "qwen2",
|
||||
"vocab_size": 151936,
|
||||
"hidden_size": 1536,
|
||||
"intermediate_size": 8960,
|
||||
"num_hidden_layers": 28,
|
||||
"num_attention_heads": 12,
|
||||
"num_key_value_heads": 2,
|
||||
"hidden_act": "silu",
|
||||
"max_position_embeddings": 131072,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_theta": 10000.0,
|
||||
"tie_word_embeddings": false,
|
||||
"torch_dtype": "bfloat16",
|
||||
"use_cache": true,
|
||||
"bos_token_id": 151646,
|
||||
"eos_token_id": 151643
|
||||
}
|
||||
Reference in New Issue
Block a user