atcsecure (@atc_secure) 's Twitter Profile
atcsecure

@atc_secure

Technology Guru, Blockchain enthusiast, decentralizing the world, one bit at a time...

ID: 24745006

linkhttps://blocknet.co/ calendar_today16-03-2009 18:47:49

456 Tweet

677 Followers

1,1K Following

atcsecure (@atc_secure) 's Twitter Profile Photo

Autonomous Agents at work... takes a single prompt, designs, documents, builds and debugs from start to finish, and supports iterative steps, which enables the ability to take over from a previous dev (human or ai based), Supports JIRA integration, github, bitbucket, etc....

Autonomous Agents at work... takes a single prompt, designs, documents, builds and debugs from start to finish, and supports iterative steps, which enables the ability to take over from a previous dev (human or ai based), Supports JIRA integration, github, bitbucket, etc....
Alvaro Cintas (@dr_cintas) 's Twitter Profile Photo

The new Anthropic prompt engineering tool is incredible. You just need to write your goal and Claude will generate an optimized prompt instantly. Here is how to use it:

atcsecure (@atc_secure) 's Twitter Profile Photo

Will be interesting to see if these results are easily duplicated... as I've seen a lot of gemini users report sub-par results

atcsecure (@atc_secure) 's Twitter Profile Photo

With so many models, now open source vision models... although I don't any 70b vision models.... I wonder how a FrankenMOE with vision would yield

LlamaIndex 🦙 (@llama_index) 's Twitter Profile Photo

Watch the benefits of Google Gemini's gigantic 1 million token context window in action! In this quick notebook, we show Gemini built into a LlamaIndex agent attempting to answer a multi-part question from a set of complicated, heterogeneous documents. The longer the context

Watch the benefits of <a href="/Google/">Google</a> Gemini's gigantic 1 million token context window in action!

In this quick notebook, we show Gemini built into a LlamaIndex agent attempting to answer a multi-part question from a set of complicated, heterogeneous documents. The longer the context
atcsecure (@atc_secure) 's Twitter Profile Photo

The dolphin-2.9.2-qwen2-72b model from Eric Hartford --- is very impressive, by far the best opensource model I have used... (exl2: 8b quant: huggingface.co/atcsecure/dolp…) Checkout this response, specifically the last paragraph

The dolphin-2.9.2-qwen2-72b model from <a href="/cognitivecompai/">Eric Hartford</a>  --- is very impressive, by far the best opensource model I have used... (exl2: 8b quant: huggingface.co/atcsecure/dolp…)

Checkout this response, specifically the last paragraph
atcsecure (@atc_secure) 's Twitter Profile Photo

llama 3.1-70b - eval using a classic Matthew Berman question did meta add a sensor of humor? ```Thanks for keeping me grounded (pun intended)!```

llama 3.1-70b - eval using a classic <a href="/MatthewBerman/">Matthew Berman</a> question

did meta add a sensor of humor?
```Thanks for keeping me grounded (pun intended)!```