Eric Buehler
Eric Buehler
I am trying to use `topk` to implement X-LoRA in Candle, and want to perform `topk` in the last dimension. Specifically, I need the `indices` return value (as returned by...
Paper link: https://arxiv.org/abs/2402.07148 This PR integrates [X-LoRA](https://github.com/EricLBuehler/xlora) by creating a new tuner model type on the level of `LoraModel`. Please see #1472. ## Changes Although the new model type is...
Hello all, Thank you for your excellent work here! We are using `Tokenizer::from_file` to load the `tokenizer.json` file from HF hub. However, it produces many warnings when loading the Phi3...
Hello all, We are implementing the Idefics 2 model on mistral.rs, but the HF Transformers code [here](https://github.com/huggingface/transformers/blob/c96aca3a8d66d64f868a3e3967be624d79213bef/src/transformers/models/idefics2/modeling_idefics2.py#L181-L182) uses `torch.bucketize` as a critical part of the code. Is it possible to...
Hello @mokeyish, Thanks for your excellent work here. We are implementing Idefics 2 in mistral.rs, but the Hugging Face implementation we are basing it on uses `torch.bucketize` and `torch.unfold`. Given...
This PR adds support for our first multimodal model: Idefics 2 (https://huggingface.co/HuggingFaceM4/idefics2-8b)! **Implementation TODOs:** - [x] VisionTransformer - [x] Encoder - [x] Attention - [x] MLP - [x] VisionEmbeddings (pending...
If this works, we can extend it to the other models. Hopefully, this will fix the problem in #339 for models without sliding window attention.
This PR implements our first embedding model: nomic-ai/nomic-embed-text-v1!
- [ ] Support for LongRope (this is supported with ISQ in non-GGUF models, though) - The challenge is that the scalings information is not present in the GGUF file....