minus-squareSuper_Pole_Jitsu@alien.topBtoLocalLLaMA•Your settings are (probably) hurting your model - Why sampler settings matterlinkfedilinkEnglisharrow-up1·1 year agoThis is absolutely golden, and is probably the reason for the absolutely shit performance I got on my local models. You should definitely write a paper about this! linkfedilink
minus-squareSuper_Pole_Jitsu@alien.topBtoLocalLLaMA•What is the best code generation model aside from gpt-4?linkfedilinkEnglisharrow-up1·1 year agoThe phind.com model seems decent linkfedilink
This is absolutely golden, and is probably the reason for the absolutely shit performance I got on my local models. You should definitely write a paper about this!