Update README.md
Browse files
README.md
CHANGED
|
@@ -46,5 +46,5 @@ https://github.com/ggerganov/ggml/issues/745
|
|
| 46 |
* https://github.com/twenkid
|
| 47 |
|
| 48 |
...
|
| 49 |
-
* Other Bulgarian
|
| 50 |
The company didn't live long, it was a show-off. Now it seems reasonable that their model was GPT2-SMALL, as that was the usual choice even 4 years later and even the Bulgarian Academy of Science 2023 model was the small one. I found several other GPT2-SMALL models trained later than this one here, one for poetry, the BAS' from 2023 and maybe a few others. I couldn't get info from the ML engineer of the BAIHUI project M.V.
|
|
|
|
| 46 |
* https://github.com/twenkid
|
| 47 |
|
| 48 |
...
|
| 49 |
+
* Other Bulgarian autoregressive models: an earlier one was a few seconds display of a generation in Bulgarian by a startup called BAIHUI AI in mid 2019. I've written in my blog 1.5B, but I don't remember if they have mentioned a size and now it seems unlikely and unreasonable, they just showed that they can train a model, a team of 3 people, only one of them a ML engineer. There are a few surviving records: my blog post: https://artificial-mind.blogspot.com/2019/07/baihuiai-baihuiai-new-bulgarian-ai.html and info here: https://www.eu-startups.com/directory/baihui-ai/
|
| 50 |
The company didn't live long, it was a show-off. Now it seems reasonable that their model was GPT2-SMALL, as that was the usual choice even 4 years later and even the Bulgarian Academy of Science 2023 model was the small one. I found several other GPT2-SMALL models trained later than this one here, one for poetry, the BAS' from 2023 and maybe a few others. I couldn't get info from the ML engineer of the BAIHUI project M.V.
|