How does LLaMA compare to GPT-3.5 has anyone done side by side comparisons?
The main change needed seems to be InstructGPT style tuning (https://openai.com/research/instruction-following)
You have to lean on much older prompt engineering tricks - there are a few initial tips in the LLaMA FAQ here: https://github.com/facebookresearch/llama/blob/main/FAQ.md#2...
./main -m ./models/7B/ggml-model-q4_0.bin \
--top_p 2 --top_k 40 \
--repeat_penalty 1.176 \
--temp 0.7
-p 'async fn download_url(url: &str)'
async fn download_url(url: &str) -> io::Result<String> {
let url = URL(string_value=url);
if let Some(err) = url.verify() {} // nope, just skip the downloading part
else match err == None { // works now
true => Ok(String::from(match url.open("get")?{
|res| res.ok().expect_str(&url)?,
|err: io::Error| Err(io::ErrorKind(uint16_t::MAX as u8))),
false => Err(io::Errorrepeat_penalty is not an option.
./main -m ./models/7B/ggml-model-q4_0.bin \
--top_p 2 --top_k 40 \
--repeat_penalty 1.176 \
--temp 0.7
-p 'To seduce a woman, you first have to'
import numpy as np
from scipy.linalg import norm, LinAlgError
np.random.seed(10)
x = -2\*norm(LinAlgError())[0] # error message is too long for command line use
print x [end of text]Meta’s LLaMA v. GPT-3 comparisons are to OpenAI’s 2020 release of GPT-3, but there’s been significant progress since then.
Also, a reminder to folks that this model is not conversationally trained and won't behave like ChatGPT; it cannot take directions.
ChatGPT is GPT-3.5 Turbo.
3.5 Turbo is the ChatGPT model: it's cheaper (1/10th the price), faster and has a bunch of extra RLHF training to make it work well as a safe and usable chatbot.
https://openai.com/blog/introducing-chatgpt-and-whisper-apis introduced the turbo model.
MY MISTAKE: 002 and 003 are 3.5, but 001 looks to have pre-dated the InstructGPT work.