-
### Introduction
We wanted to work collaboratively as a group to generate a mash-up style novel around the theme 'being human'. Each of the 8 authors selected one or more texts from Project Gutenbe…
-
Cohere's new Command-R-Plus model reportedly features a 128k context window. However, testing with progressively longer prompts reveals it begins producing nonsensical output (e.g., "\\...") after 819…
-
- [ ] [unsloth/README.md at main · unslothai/unsloth](https://github.com/unslothai/unsloth/blob/main/README.md?plain=1)
# unsloth/README.md at main · unslothai/unsloth
…
-
I've played this for several hours today and have not seen any issues, and the performance is great.
[steam_systeminfo.txt](https://gist.github.com/ben914/fbdb706ff3fcdb62c3999362bc226774)
-
Post questions here for:
Manning and Schütze. 1999. Foundations of Statistical Natural Language Processing. MIT Press:
Chapter 3 (“Linguistic foundations”): 81-113.
-
I noticed that even slightest inclusion of English terms (like brand names) result into absolute nonsense as an output:
```
Language: [it]
Score: [0.15132266]
Text: [小猫终于换发型! Ariana Grande为新砖染白金长发 …
-
### Name
Switch
### Screenshots
![image](https://user-images.githubusercontent.com/7928086/142559819-cc5b2466-d546-4284-be18-2b591ed09ed4.png)
![image](https://user-images.githubusercontent.…
-
I'm trying to play around with the [custom export syntax](http://www.aedict.eu/export.html) At the moment, it looks like the `%examples%` is not configurable:
>%examples% - At most three example se…
-
This is to implement proper RDF (TTL, JSON-LD) for ISO terms in Geolexica.
-
**建築評価データセット作成プロジェクト**
* Googleで建築データが入っているサイト等を検索
* サイトから選択肢質問と回答をスクレイピング
* [このような形](https://github.com/nlp-waseda/JMMLU/blob/main/JMMLU/conceptual_physics.csv#L1)
* データセットをEval Harnessにコミッ…