Li Jiang
Li Jiang
I am struggling with the same issue.
When I followed this instruction and install Mujoco_py. There still comes with this error, and I think it comes to a new version of M1? > ImportError: dlopen(/opt/anaconda3/envs/jl/lib/python3.8/site-packages/mujoco_py/generated/cymj_2.1.2.14_38_macextensionbuilder_38.so, 0x0002): tried:...
Thanks @vaibhavad 1. Does LLM2Vec support for other models, e.g., Pythia? I have checked it and it seems like it only supports llama and mistral families. 3. Just to confirm:...
I've updated the reference. Please check on it.
I have the same concern,.
Thanks! How to determine what embedding way we should use? I think we should always use `bbcc` for embedding (embedding tasks require bidirectional attention). Please clarify me if I am...
Sorry, I think I am a little bit confused. If my training code is `mode= unified`, what is the suggestion for `attn`?
Thanks! I still have three questions here: 1. Just to confirm, the paper said that there are two things (Figure 3): a. prompt design for embedding tasks. b. uses bidirectional...
I just reviewed this problem and found similar issues in #24 and #15 regarding `attn=cccc`. However, I find that they are specific to the `mistral` model, and how about other...