nightmedia commited on
Commit
b166fc6
·
verified ·
1 Parent(s): 6f70959

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +0 -3
README.md CHANGED
@@ -15,9 +15,6 @@ library_name: mlx
15
 
16
  Hi Spock!
17
  We are going to analyze the cognitive abilities of a few quantizations of this model
18
- - The bf16 is full precision.
19
- - The q6 is straight quantization with the MLX default settings (group size 64)
20
- - The hi quants are done with group size 32 for higher fidelity
21
 
22
  The Deckard(qx) quants are in a mixed precision quantization:
23
  - qx64x has data at 4 bit, while the attention paths, head, and embeddings are at 6 bit
 
15
 
16
  Hi Spock!
17
  We are going to analyze the cognitive abilities of a few quantizations of this model
 
 
 
18
 
19
  The Deckard(qx) quants are in a mixed precision quantization:
20
  - qx64x has data at 4 bit, while the attention paths, head, and embeddings are at 6 bit