search

Xatta-Trone / medium-parser-extension

Read medium.com and medium based articles using google web cache.
MIT License
1.25k stars 76 forks source link

Amazing work #5

Closed MeWannaSleep closed 11 months ago

MeWannaSleep commented 11 months ago

Thx for your amazing work. but reading from the google cache, I found some contents are not properly displayed, image I don't know whether it's because the data from google cache is broken or something else went wrong the medium link is https://towardsdatascience.com/fine-tune-your-own-llama-2-model-in-a-colab-notebook-df9823a04a32

Xatta-Trone commented 11 months ago

I see why it happened. The Google web cache does not include the hosted code snippets. That is why I have added archive.is as a fallback option.

In your case, in the main code snippet

result = pipe(f"<s>[INST] {prompt} [/INST]") In here, the <s> tag caused this issue. Because in the extension, it is parsing the HTML contents only.

You can always revert to archive.is as a fallback using the Read from Archive.is button.

image

In the archive.is it is working okay. Here is the link to URL https://archive.is/KrD7I#selection-1933.0-1941.1

@MeWannaSleep

MeWannaSleep commented 11 months ago

thx for your reply,