Files
Apertus-3B-DPO-wnorm2both/config.json

35 lines
784 B
JSON
Raw Permalink Normal View History

{
"architectures": [
"ApertusForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 1,
"dtype": "bfloat16",
"eos_token_id": 68,
"hidden_act": "xielu",
"hidden_dropout": 0.0,
"hidden_size": 3072,
"initializer_range": 0.02,
"intermediate_size": 16384,
"max_position_embeddings": 4096,
"mlp_bias": false,
"model_type": "apertus",
"num_attention_heads": 24,
"num_hidden_layers": 24,
"num_key_value_heads": 8,
"pad_token_id": 10,
"post_norm": false,
"qk_norm": true,
"rms_norm_eps": 1e-05,
"rope_parameters": {
"rope_theta": 500000.0,
"rope_type": "default"
},
"rope_theta": 500000.0,
"tie_word_embeddings": false,
"transformers_version": "5.7.0",
"use_cache": false,
"vocab_size": 131072
}