DirectionOdd9824

joined 10 months ago
sorted by: new top controversial old
Anyone have a 1B or 3B model that is mostly coherent? in c/localllama@poweruser.forum
[โ€“] DirectionOdd9824@alien.top 1 points 10 months ago

Did any try phi-1.5?

Options for fine-tuning, advice please in c/localllama@poweruser.forum
[โ€“] DirectionOdd9824@alien.top 1 points 10 months ago

you get 14hrs of a100 80gb with 25 dollars.

1
How to finetune llama 2 on a tpu? (alien.top)
 

I'm trying to train llama 2 on a tpu using qlora and peft. All the scripts I find are tied to CUDA. Are there any available?