d-matrix commited on
Commit
5869676
1 Parent(s): e6c0785

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -3
README.md CHANGED
@@ -36,13 +36,18 @@ model-index:
36
  type: dmx-perlexity
37
  value: 46.570838928222656
38
  ---
39
- This is a d-Matrix functional reference of [`gpt2-xl`](https://huggingface.co/openai-community/gpt2-xl).
40
-
41
- It provides the following configurations:
 
 
 
42
 
 
43
  - `BASELINE`: a reference functionally equivalent to the original model.
44
  - `BASIC`: all linear algebraic operations quantized to `BFP16-64`, and all other operations transformed to corresponding functional kernel simulations.
45
 
 
46
  ### Usage Example
47
 
48
  Prerequisites:
 
36
  type: dmx-perlexity
37
  value: 46.570838928222656
38
  ---
39
+ This is a d-Matrix functional reference of the GPT2 model family, of the following revisions:
40
+ - [`distilgpt2](https://huggingface.co/distilbert/distilgpt2)
41
+ - [`gpt2`](https://huggingface.co/openai-community/gpt2)
42
+ - [`gpt2-medium`](https://huggingface.co/openai-community/gpt2-medium).
43
+ - [`gpt2-large`](https://huggingface.co/openai-community/gpt2-large).
44
+ - [`gpt2-xl`](https://huggingface.co/openai-community/gpt2-xl) (default).
45
 
46
+ The reference provides the following functional configurations:
47
  - `BASELINE`: a reference functionally equivalent to the original model.
48
  - `BASIC`: all linear algebraic operations quantized to `BFP16-64`, and all other operations transformed to corresponding functional kernel simulations.
49
 
50
+
51
  ### Usage Example
52
 
53
  Prerequisites: