lynx

I dont know what you mean with steering?

• Do you want a given output structure, like json or toml?
• Do you want to align the model, with your dataset of question and answer pairs?

First of all, have you tried giving the model multiple examples of input output pairs in the context, this already helps the model a lot to output the correct format.

Second you can force a specific output structure by using a regex or grammar: https://python.langchain.com/docs/integrations/chat/outlines/#constrained-generation https://github.com/ggerganov/llama.cpp/blob/master/grammars/README.md

And third, in case you want to train a model to respond differently and the previous steps were not good enough, you can fine-tune. I can recommend this project to you, as it teaches how to fine-tune a model: https://github.com/huggingface/smol-course

Depending on the size of the model, that you want to fine-tune and the amount of compute that you have available you can either train by updating all parameters like ORPO or you can train via PEFT (LoRA)

First of all i think it is a great idea to give the model access to a map. Unfortunately it seems like, that the script is missing a huge part at the end, the loop does not have any content and the Tools class is missing.

I have found the problem with the cut off, by default aider only sends 2048 tokens to ollama, this is why i have not noticed it anywhere else except for coding.

When running /tokens in aider:

$ 0.0000   16,836 tokens total
           15,932 tokens remaining in context window
           32,768 tokens max context window size

Even though it will only send 2048 tokens to ollama.

To fix it i needed to add a file .aider.model.settings.yml to the repository:

- name: aider/extra_params
  extra_params:
    num_ctx: 32768

If you want in line completions, you need a model that is trained on "fill in the middle" tasks. On their Huggingface page they even say that this is not supported and needs fine tuning:

We do not recommend using base language models for conversations. Instead, you can apply post-training, e.g., SFT, RLHF, continued pretraining, etc., or fill in the middle tasks on this model.

A model that can do it is:

• starcoder2
• codegemma
• codellama

Another option is to just use the qwen model, but instead of only adding a few lines let it rewrite the entire function each time.

Split Horizon with Poison Reverse

This is probably the only reason microsoft recall exists, as it is completely useless for anything else.

On Huggingface is a space where you can select the model and your graphics card and see if you can run it, or how many cards you need to run it. https://huggingface.co/spaces/Vokturz/can-it-run-llm

You should be able to do inference on all 7b or smaller models with quantization.

Thanks for the suggestion, I tried it and the diff view is very good. The setup was not really easy for my local models, but after i set it up, it was really fast. The biggest problem with the tool is that the open source models are not that good, i tried if it could fix a bug in my code and it was only able to make it worse. On a more positive note, you at least do not need to copy all text over to another window and it is great for generating boilerplate code nearly flawlessly every time.

Question: What is the best self hosted coding assistant?

The (only) project i found, that does what i want:

• https://github.com/fauxpilot/fauxpilot
• https://github.com/Venthe/vscode-fauxpilot

It works ok for the most part. The problem i have with it is that inline completion is more annoying then helpful, because the AI only sees the last few lines that you wrote and therefore does not know the larger context of the project.

I also found this project, it looks promising. Has anyone tested it? Can you separate the server from the client?

• https://github.com/morph-labs/rift

Are there other projects that integrate well into an IDE?

It is https://github.com/searx/searx but you should use searxng instead: https://github.com/searxng/searxng

the emotion is very human

So this would be the same but a bit more clear.

Solve 8+9 by creating blocks that sum up to 10. Then add the rest on top.