yogthos

joined 4 years ago
MODERATOR OF
[–] [email protected] 2 points 8 hours ago

Yeah, it's a good series overall.

[–] [email protected] 3 points 9 hours ago

It is rare that you run a book that sucks you in the way Piranesi does.

[–] [email protected] 3 points 9 hours ago (2 children)

It left a big impression on me as well, the world the way he sees it is so peaceful and tranquil, but then you start gradually realizing the horrific situation he's actually in. And this contrast between the way the character perceives his circumstances and the reality of the situation is kind of haunting.

[–] [email protected] 2 points 13 hours ago (4 children)

Piranesi is a real gem, I ran across it last year and it was absolutely delightful.

[–] [email protected] 3 points 13 hours ago (2 children)

a few books that I found enjoyable recently

  • Doors of Sleep
  • The City and the Stars
  • The Windup Girl
  • Consider Phlebas
  • A Scanner Darkly
  • The Lifecycle of Software Objects
  • The Mountain in the Sea
 
[–] [email protected] 4 points 23 hours ago

So when Estonian jails fill up, they'll ship them as volunteers to Ukraine?

[–] [email protected] 5 points 23 hours ago

You know libs, when the butcher of Iraq throws his support for your candidate, maybe it's time to accept that you are the baddies.

[–] [email protected] 3 points 23 hours ago

The discussion here is regrading the behavior of the US government when a US citizen was detained by the client regime of the US. These are the services that US would normally offer that were denied in this case https://travel.state.gov/content/travel/en/international-travel/emergencies/arrest-detention.html

[–] [email protected] 3 points 1 day ago

Nice straw man you got there. Nowhere did I say that American law applies to foreign countries. What was actually said that the law he broke in Ukraine was one restricting freedom of expression. A supposed value that US is helping Ukraine defend. Not only that, but apparently US regime didn't really care about the fact that their puppet regime in Ukraine arrested and tortured a US citizen for exercising his speech, the core value that US regime purports to be fighting for. Quite telling that you're aggressively working not to understand what's happening here.

[–] [email protected] 3 points 1 day ago (2 children)

It's interesting that you apparently support the Ukrainian censorship of freedom of expression applied to US citizens. Very revealing.

 
 
 
 

"Reflection 70B holds its own against even the top closed-source models (Claude 3.5 Sonnet, GPT-4o).

It’s the top LLM in (at least) MMLU, MATH, IFEval, GSM8K.

Beats GPT-4o on every benchmark tested.

It clobbers Llama 3.1 405B. It’s not even close.

The technique that drives Reflection 70B is simple, but very powerful.

Current LLMs have a tendency to hallucinate, and can’t recognize when they do so.

Reflection-Tuning enables LLMs to recognize their mistakes, and then correct them before committing to an answer.

Additionally, we separate planning into a separate step, improving CoT potency and keeping the outputs simple and concise for end users.

Important to note: We have checked for decontamination against all benchmarks mentioned using @lmsysorg’s LLM Decontaminator.

The weights of our 70B model are available today on @huggingface here: https://huggingface.co/mattshumer/Reflection-70B

@hyperbolic_labs API available later today.

Next week, we will release the weights of Reflection-405B, along with a short report going into more detail on our process and findings.

Most importantly, a huge shoutout to @csahil28 and @GlaiveAI.

I’ve been noodling on this idea for months, and finally decided to pull the trigger a few weeks ago. I reached out to Sahil and the data was generated within hours.

If you’re training models, check Glaive out.

This model is quite fun to use and insanely powerful.

Please check it out — with the right prompting, it’s an absolute beast for many use-cases.

Demo here: https://reflection-playground-production.up.railway.app/

405B is coming next week, and we expect it to outperform Sonnet and GPT-4o by a wide margin.

But this is just the start. I have a few more tricks up my sleeve.

I’ll continue to work with @csahil28 to release even better LLMs that make this one look like a toy.

Stay tuned."

https://x.com/mattshumer_/status/1831767014341538166

view more: next ›