Yes, def, did you see the snapdragon keynote? 30 tokens a second for Llama2 7B param on device. Maximum 13B param available to run on device. Apple can def run LLM’s on M2/M3
LLMS
-
Language Models Enable Increasingly Autonomous Systems Development
By
–
ICYMI: Earlier this month, Stanford assistant professor @nickhaber
’s seminar highlighted some of the opportunities language models provide for us to make increasingly autonomous systems. Watch the recording: https://
bit.ly/3sb0drl -
Pietro’s Magical Work in Fine-Tuning Space
By
–
Pietro is doing some absolutely magical work in the fine-tuning space. Make sure you stay up to date with everything he does. H/t @skirano
-
8-Hour GPT-4 Conversation Yields 68 Pages Transcripts
By
–
This is me on my drive between Sweden and Spain, spent probably 8 hours chatting with GPT4 Some absolute amazing outputs from the conversations we had. Roughly 68 pages of full transcripts. Hard to wrap my head around it.
-
DataRobot Leads LLMOps Market with Generative and Predictive AI
By
–
@CBinsights highlights the Large Language Model Operations (#LLMOps) market map, featuring DataRobot in the critical role of the #AI platform. The DataRobot AI Platform uniquely brings both generative and predictive AI capabilities together, delivering an open and end-to-end
-
DataRobot Leads LLMOps Market with Generative and Predictive AI
By
–
@CBinsights highlights the Large Language Model Operations (#LLMOps) market map, featuring DataRobot in the critical role of the #AI platform. The DataRobot AI Platform uniquely brings both generative and predictive AI capabilities together, delivering an open and end-to-end
-
Google Gemini Outperforms GPT-4 by 5X in Latest Benchmarks
By
–
Google Gemini Eats The World – Gemini Smashes GPT-4 By 5X, The GPU-Poors https://
bit.ly/3PcLUtK
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
IBM Granite 13B LLM Training Datasets Revealed
By
–
datasets that IBM used to train its granite.13b LLM: Arxiv, Common Crawl, DeepMind Mathematics, Free Law, GitHub Clean, Hacker News, OpenWeb Text, Project Gutenberg, Pubmed Central, SEC filings, Stack Exchange, USPTO, Webhose, Wikimedia https://
ibm.com/downloads/cas/
X9W4O6BM
… -
End-to-End Spoken Language Model Processing Spectrograms Directly
By
–
Introducing a new approach for spoken language modeling trained end-to-end to directly process spectrograms as both input and output. It can be fine-tuned to generate semantically accurate spoken language for continuation and question answering. → https://
goo.gle/3FzJNvy -
5 Game-Changing GPT-4 Applications Without Coding Skills
By
–
(new video!) 5 Game-Changing Applications of GPT-4: No Coding Skills Required! Learn more: https://
youtu.be/lwNy4lgDpjY #gpt #gpt4 #chatgpt #llama2 #openai #whatsai #ai