jeffra commited on
Commit
b9e8309
·
verified ·
1 Parent(s): d4e4797

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +25 -0
config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ArcticLSTMSpeculatorPreTrainedModel"
4
+ ],
5
+ "base_model_name_or_path": "meta-llama/Llama-3.3-70B-Instruct",
6
+ "input_hidden_dim": 8192,
7
+ "inner_dim": "6144",
8
+ "proj_dim": "6144",
9
+ "emb_dim": "6144",
10
+ "model_type": "mlp_speculator",
11
+ "n_candidates": 3,
12
+ "n_predict": 3,
13
+ "scale_input": true,
14
+ "tie_weights": true,
15
+ "tie_lstm_embs": true,
16
+ "top_k_tokens_per_head": [
17
+ 1,
18
+ 1,
19
+ 1
20
+ ],
21
+ "torch_dtype": "bfloat16",
22
+ "transformers_version": "4.47.0",
23
+ "vocab_size": 128256,
24
+ "method": "sum_lstm"
25
+ }