Alazar
Alazar
Hi @Rocketknight1! A few things: 1. Caching isn't implemented yet and so i am skipping tests that test that, please look at [this commit](https://github.com/huggingface/transformers/pull/29400/commits/a0660efd40c485e7c2d2fddc12630d846f056620) 2. Since caching isn't implemented yet,...
Hi @Rocketknight1 - I meant I am not planning on implementing the StaticCache so `past_key_values` will always be None. Yes we are definitely able to return and accept some kind...
Hi @Rocketknight1 - Unrelated to this but why is it that I can't find openELM source code in the Github repo but I see it in the hub https://huggingface.co/apple/OpenELM-270M/blob/main/modeling_openelm.py?
Closing PR in favor of KerasNLP being able to load HF models!
@merronmuche You will be porting this :)
Hi @Rocketknight1! I tried creating a gpt4 draft using the prompt you shared with me but it wasn't translating all the classes, not sure why, did you have to translate...
Hi @Rocketknight1 - Good to hear from you! Thank you for letting me know. One reason I was adding TF models to Transformers was so i can then easily (optimize...
Sounds good @Rocketknight1! Let me play with the KeradNLP api and get back to you. On a different note, are there any beginner level bugs you can point me too?...
closing this PR in favor of keras-nlp
Hello @SunMarc @amyeroberts - if nobody started on this, i would like to contribute to this by adding the GGUF for phi3, is following Qwen integration as a guide good...