File size: 439 Bytes
a9469a5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
model_info:
  name: anemll-Meta-Llama-3.2-3B-ctx512
  version: 0.1.1
  description: |
    Demonstarates running Meta-Llama-3.2-3B on Apple Neural Engine
    Context length: 512
    Batch size: 64
    Chunks: 2
  license: MIT
  author: Anemll
  framework: Core ML
  language: Python
  parameters:
    context_length: 512
    batch_size: 64
    lut_embeddings: none
    lut_ffn: 4
    lut_lmhead: 4
    num_chunks: 2
    model_prefix: llama