A fully uncensored Hermes 3 with lorablation! The lorablated model directly answers questions without any tweaking. Here's how it was made: 1/ Create a LoRA adapter based on Llama 3.1 8B Instruct and my abliterated version. 2/ Apply it to Hermes 3 from @NousResearch using
LLMS
-
LLM Chatbots vs Agentic Systems: Decision-Making Capabilities
By
–
Really interesting, @sanjaykalra
. This part tho… it gives me pause: "While LLM chatbots have no ability to make actual decisions, agentic systems are characterized by advanced contextual reasoning and decision-making. Agentic systems can plan, “understand” intent, and more -
User reports improved factual accuracy in Gemini model
By
–
I love the new @GeminiApp model! Much more factually accurate than before AI Optimisation works!
-
Tamper-Resistant Safeguards for Open-Weight LLMs
By
–
Excited to feature Tamper-Resistant Safeguards for Open-Weight LLMs from @lapisrocks
! Introducing the first safeguards for LLMs that resist fine-tuning attacks, showing the power of tamper-resistance to make open-weight LLMs safer. @rishub_t is here to answer your questions! -
MergeKit and PruneMe: Essential Tools for Model Optimization
By
–
No, I would recommend libraries like MergeKit or PruneMe (
https://
github.com/arcee-ai/Prune
Me
…) from @arcee_ai for pruning. -
Live Fine-Tuning Models Session with DataCamp and Google Colab
By
–
If you want to start fine-tuning models, I'll do a live session on August 27 with @DataCamp using @GoogleColab
! -
Claude Artifacts adds support for LaTeX rendering
By
–
I just realised that Claude seems to be supporting Latex too within Artifacts
-
Grok 2.0 Mini Adds Support for LaTeX Formulas
By
–
ICYMI: Grok 2.0 mini supports LaTeX formulas https://
x.com/yacineMTB/stat
/yacineMTB/status/1824135330695663791
… -
Survey on Prompt Engineering Strategies and Techniques
By
–
If you’re looking to learn more about prompt engineering, here’s a good survey paper that outlines the different types of prompt engineering (link provided in alt) Here are some of the strategies discussed:
•Chain of Thought (CoT)
•Chain of Cause (CoC)
•Program-Aided Language -
Modular Speech-to-Speech Pipeline Achieves Sub-500ms Latency
By
–
this modular speech-to-speech pipeline repo with <500 ms second latency (on a local GPU) is going crazy
— Thomas Wolf (@Thom_Wolf) 16 août 2024
Impressive what a well tuned pipeline of stt-llm-tts can do
cool work by @eustachelb with @reach_vb and @andi_marafioti
find the code here: https://t.co/HrbqSWQ3sj pic.twitter.com/HDkEnVntKNthis modular speech-to-speech pipeline repo with <500 ms second latency (on a local GPU) is going crazy Impressive what a well tuned pipeline of stt-llm-tts can do cool work by @eustachelb with @reach_vb and @andi_marafioti find the code here: https://
github.com/eustlb/speech-
to-speech
…