ChatGPT seems to be testing an updated sidebar layout where the "New Chat" button is replaced with "Search". A list of pinned GPTs is also hidden, and the sidebar is fully dedicated to the conversation history itself
LLMS
-
Logit-Viewer: Visualize Language Model Probability Distributions
By
–
logit-viewer This python script provides a user-friendly Gradio interface for inspecting logits produced by the forward pass of language models after applying the softmax function. It allows you to input text and visualize the probability distribution over the vocabulary for
-
Getting Started with OpenAI o1 Reasoning Models Guide
By
–
Getting Started with OpenAI o1 Reasoning Models: Learn how to use the OpenAI o1-preview & o1-mini for decision-making, coding, and building an end-to-end machine learning project from scratch. https://
kdnuggets.com/getting-starte
d-with-openai-o1-reasoning-models?utm_source=dlvr.it&utm_medium=twitter&utm_campaign=getting-started-with-openai-o1-reasoning-models
… -
Personal challenges with AI writing and prompt engineering
By
–
100% – I spent a year trying to prompt a good writing style and am still not there yet – 40% of my texts I rather have to dictate than delegate to AI.
-
SambaNova Cloud Delivers Fastest Llama 3.1 Inference
By
–
We've spoken with the devs and they've told us what they need: fast inference. ⚡️
— SambaNova (@SambaNovaAI) 30 septembre 2024
SambaNova Cloud delivers the fastest inference on @AIatMeta's Llama 3.1 405B, 70B and 8B — all running at full precision.
Start developing 👉 https://t.co/zm6RCXY00n#API #AI @aton2006 pic.twitter.com/ORu1lalYAhWe've spoken with the devs and they've told us what they need: fast inference. SambaNova Cloud delivers the fastest inference on @AIatMeta
's Llama 3.1 405B, 70B and 8B — all running at full precision. Start developing http://
cloud.sambanova.ai #API #AI @aton2006 -
Implementing AI automation with GPTs and custom functions
By
–
That would be very nice but feels like we are a way too far from this for now. It is kind of possible with GPTs and custom functions already though You just have to build events tracking by yourself
-
Cerebras Achieves 2000 Tokens Per Second Llama Inference
By
–
4/6 Daniel Kim, Head of DevRel at Cerebras, gave a behind the scenes talk on how to support over 2000 tok/s inference with llama models.
-
Llama Ecosystem Panel with Swyx, Jerry Liu, and Angela Yeung
By
–
5/6 We also had an exceptional panel, with Swyx, editor of @latentspacepod
, Jerry Liu, founder of @llama_index
, and Angela Yeung, VP of product at Cerebras, discussing the overall llama landscape and what's coming next. -
Superhuman Launches Ask AI Assistant Powered by Llama
By
–
3/6 Roland Garvilescue, lead AI at @Superhuman
, shared Ask AI, an AI-assistant inside Superhuman that's powered by llama. -
Llama Models Surpass 350 Million Downloads on Hugging Face
By
–
Llama models have over 350 million downloads on hugging face! Last week, we gathered this massive llama community at Llamapalooza, a one-of-a-kind event all about fine-tuning and using llama in production.