I like seeing a snippet of code!
@simonw
-
Running Meta-Llama-3-8B-Instruct with LLM CLI Tool
By
–
Yes, you can run prompts like this: llm -m Meta-Llama-3-8B-Instruct -o temp 0.1 "3 names for a pet rhino" https://
github.com/simonw/llm-gpt
4all?tab=readme-ov-file#model-options
… -
Evaluating Speed Cost and Offline AI Performance Metrics
By
–
Speed, cost and the ability to run offline on a personal device are the things I care most about I still don't have a great answer for how to best evaluate these things but I'm researching that at the moment
-
Prompt Injection vs Jailbreaking: Understanding AI Security Distinctions
By
–
prompt injection is a security issue – you may be confusing it with jailbreaking https://
simonwillison.net/2024/Mar/5/pro
mpt-injection-jailbreaking/#censorship-debate
… -
GPT-3.5 prompt injection defense and fine-tuning advantages
By
–
I quite like that they're using GPT3.5 here – for prompt injection a defense that works with the less expensive, faster models would almost certainly work with the more expensive ones too I imagine GPT-3.5 is better for fine tuning at the moment because there's more experience
-
Real-world AI attacks remain limited beyond screenshots and prompt leaks
By
–
There have been a bunch of research/demo attacks, but I haven't seen a real-world malicious attack that caused actual damage yet – beyond producing embarrassing screenshots or leaking system prompts
-
51 Posts on Prompt Injection: Complete Collection
By
–
Here are all 51 of my posts on prompt injection so far – love having tags on my blog!
-
Streaming API Usage JSON Bug Report and Feature Request
By
–
That's great, thanks! I'll continue to hold out on my dream for daily spending limits – I may have to implement that myself, though it would be easier if you fix the bug where streaming responses don't get valid "usage" JSON blocks
-
Fine-tuned Models Still Vulnerable to Adversarial Attacks
By
–
TLDR version: the fine-tuned model they describe improves things, but I still don't think "improves" is good enough If you are facing an adversarial attacker then reducing the chance that they might find an exploit just means they’ll try harder until they find one that works