Comment by sfriedr | Hacker News Reader

Comment by sfriedr | Hacker News Reader

Congratulation, great paper! It should have been put on HN earlier ;)

I have a few questions:

* you say (page 4): "We then perform standard instruction finetuning on the base LLaMA-7B model" Could you perhaps provide a reference to the _exact_ finetuning approach you used? I'm afraid different groups of people have a different notion of "standart" (see for example pages 131-155 from https://arxiv.org/abs/2302.08575 for various fine-tuning approaches) and without knowing exactly how fine-tuning was carried out, it can be very difficult reproduce your research and results exactly.

* the idea of using AST Sub-Tree Matching is nice. Could you please let me know which function in which file from your GitHub repository this is implemented in?

Again, great job on publishing this paper!

---

Best regards,

friederrr.org

Thanks @sfriedr We generate self-instruct data and then fine tune the base model with perplexity loss. The self-instruct data is https://github.com/ShishirPatil/gorilla/tree/main/data/apibe...

Thank you! Yes, the code can be found here: https://github.com/ShishirPatil/gorilla/tree/main/eval/eval-...

Hope this helps. Let me know if you have any follow-ups!

Awesome, thanks for letting me know!

I'm still not sure though about some nitpicky things: - do you change all the weights, or just the ones from the last layer when fine-tuning? - do you just train on the _code_ field from the JSON file with the self-instruct data, or do you also use the other fields to train (or do you use the other fields just for downstream evaluation purposes)?

I think it could be a major selling point of your paper if on Github (or in an appendix to your preprint, if you update it on arxiv), you had a section where you document the training process in detail

(whoops, this comment/questions should have been to as an answer to your other comment @shishirpatil)

Seems @shishirpatil ran out of steam answering questions. Too bad.

(Or maybe the questions were too tricky and he wasn't able to answer, heh)

Haha, was busy yesterday! Or was I? :P

Reply on news.ycombinator.com