peakji

7 ms·

1.
by peakji 2y ago
GGUF files are available on HF: https://huggingface.co/peakji/steiner-32b-preview-gguf I haven't personally used Ollama Modelfile, but I think it should be relatively easy to convert from GGUF?
2.
by peakji 2y ago
I'm new here. Just curious, why avoid Medium? Is it a Hacker News thing, or did I miss something?
3.
by peakji 2y ago
The result for Qwen2.5-32B (49.49) is using CoT prompting. Only Steiner models do not use CoT prompting. More importantly, I highly recommend to try these out firsthand (not only Steiner, but all reasoning models). You'll find that the
4.
by peakji 2y ago
It is an LLM fine-tuned using a new type of dataset and RL reward. It's good at reasoning, but I would not recommend to replace Llama for general tasks.
5.
by peakji 2y ago
Approaches like best of n sampling and majority voting are definitely feasible. But I don't recommend trying things related to CoT, as it might interfere with the internalized reasoning patterns.
6.
by peakji 2y ago
The model can already answer some tricky questions that other models (including GPT-4o) have failed to address, achieving a +5.56 improvement on the GPQA-Diamond dataset. Unfortunately, it has not yet managed to reproduce inference-time sca
7.
Show HN: Steiner – An open-source reasoning model inspired by OpenAI o1(medium.com)
83 points by peakji 2y ago|19 comments
8.
Show HN: Basaran – OpenAI-compatible streaming API for your own models(github.com)
5 points by peakji 4y ago|0 comments
9.
Show HN: Ratus – RESTful asynchronous task queue server(github.com)
3 points by peakji 4y ago|0 comments