Ergebnisse
198 ArtikelSortieren
18:08EN
Matthew Berman· 29. Juni 2023Six Language Models Compared on an Informal 2023 Prompt SetMatthew Berman · Forward FutureMatthew Berman tests coding, writing and reasoning with H2O GPT, finding a tie under a small and subjective rubric.
59:36ENMatthew Berman· 27. Juni 2023Pascal Pfeiffer on Retrieval, Open Models, Kaggle and Fine-Tuning with H2O LLM StudioMatthew Berman, Pascal Pfeiffer · Forward FutureIn a 2023 interview, H2O.ai’s Pascal Pfeiffer explains document retrieval and parameter-efficient fine-tuning, reflects on Kaggle competitions and demonstrates LLM Studio. He and Matthew Berman also consider coding assistants and unverified rumors about model architecture.
7:21ENMatthew Berman· 26. Juni 2023Testing Orca Mini 13B, an Unofficial Model Inspired by OrcaMatthew Berman · Forward FutureMatthew Berman loads a quantized Orca Mini 13B and tries coding, reasoning, arithmetic, and summarization prompts. His 2023 informal test produces several failures and a few successes.
8:17ENMatthew Berman· 22. Juni 2023Using OpenLLM to Serve Open Models Through an APIMatthew Berman · Forward FutureMatthew Berman demonstrates a local model server, its interactive API and a Python client in a 2023 walkthrough.
4:36ENMatthew Berman· 20. Juni 2023Testing WizardCoder 15B as a Local Coding AssistantMatthew Berman · Forward FutureMatthew Berman runs the 2023 WizardCoder model locally, testing small Python exercises and a longer Snake game.
9:46ENMatthew Berman· 19. Juni 2023Installing Text Generation WebUI on Windows: A 2023 Local-Model WalkthroughMatthew Berman · Forward FutureMatthew Berman installs an early version of Text Generation WebUI, resolves setup errors and explains model downloads and interface settings. The steps and packages are historical; consult the project’s current documentation for a modern installation.
59:28EN
Cognitive Revolution "How AI Changes Everything"· 16. Juni 2023PodcastFrom $500,000 to $50,000: MosaicML Made Training Your Own Model a Priced ProductModeration: Nathan LabenzJonathan Frankle and Abhi Venigalla explain who is training custom LLMs, why extraction beats chat, and how ALiBi stretches a 65k context window.
10:23ENMatthew Berman· 15. Juni 2023Testing Hosted Falcon 40B and Document Chat in h2oGPTMatthew Berman · Forward FutureMatthew Berman tries H2O.ai’s 2023 hosted Falcon 40B interface on coding, writing and reasoning prompts, then demonstrates document chat.
19:38ENMatthew Berman· 9. Juni 2023Orca: Teaching a Smaller Model with GPT-4 ExplanationsMatthew Berman · Forward FutureMatthew Berman reviews Microsoft Research’s 2023 Orca paper and its approach to training a 13-billion-parameter model on detailed teacher responses.
2:41:58EN
Lex Fridman· 8. Juni 2023Podcast · 2 StimmenZuckerberg showed up as Mark Elliott — then argued for many AIs, not oneGast: Mark Zuckerberg · Moderation: Lex FridmanOn his second Lex Fridman podcast, the Meta CEO treats a disguised first jiu-jitsu tournament, an open research model called LLaMA, and a future of many agents as the same bet: ship before it is perfect, and do not build a single god model.
10:40ENMatthew Berman· 7. Juni 2023Testing Nous Hermes 13B Locally: Mixed Reasoning and Unsafe ResponsesMatthew Berman · Forward FutureMatthew Berman compares a 2023 quantized model on hosted and local hardware, with uneven results across coding, logic and safety.
8:51ENMatthew Berman· 2. Juni 2023Falcon 40B: A 2023 Leaderboard Result and a Slow GPU TestMatthew Berman · Forward FutureMatthew Berman tries a quantized Falcon 40B Instruct checkpoint on a hosted GPU and finds useful output but slow generation.
11:42ENMatthew Berman· 31. Mai 2023Guanaco and QLoRA: A 2023 Fine-Tuning and Model TestMatthew Berman · Forward FutureMatthew Berman reviews the QLoRA paper, explains the reported memory savings and tries Guanaco 65B on an informal prompt set.
1:20:49ENCognitive Revolution "How AI Changes Everything"· 25. Mai 2023PodcastA Wafer-Scale Chip So AI Stops Waiting on AmdahlModeration: Nathan LabenzAndrew Feldman, CEO of Cerebras, tells Nathan Labenz why one 2.5-trillion-transistor die, sparse linear algebra, and a 16-machine supercomputer are a better shape than a GPU cluster.
1:11:11EN
20VC with Harry Stebbings· 17. Mai 2023Podcast · 2 StimmenEmad Mostaque on Open AI Models, National Data and the Race to DeployGast: Emad Mostaque · Moderation: Harry StebbingsThe Stability AI co-founder describes his 2023 vision for open models, healthcare applications and AI businesses, alongside concerns about a speculative boom.
11:07ENMatthew Berman· 16. Mai 2023Testing MosaicML MPT-7B Chat and StoryWriter in 2023Matthew Berman · Forward FutureMatthew Berman examines MPT-7B model variants, their licenses and a mixed set of local tests.
1:06:02EN20VC with Harry Stebbings· 15. Mai 2023Podcast · 2 StimmenYann LeCun on Open AI Models, Safety and the Future of ResearchGast: Yann Lecun · Moderation: Harry StebbingsMeta’s chief AI scientist discusses language models, research priorities and his contested views on AI risk and employment.
9:58ENMatthew Berman· 26. Apr. 2023HuggingChat’s 2023 Debut: Testing Writing, Reasoning and Python CodeMatthew Berman · Forward FutureMatthew Berman tries the early OpenAssistant-powered HuggingChat on writing, arithmetic, logic and a Snake game, then compares a few outputs with GPT-3.5 and GPT-4. The results illustrate both promise and limitations in this small historical test.
1:13:52ENMatthew Berman· 15. Apr. 2023Andriy Mulyar on GPT4All, Data Quality and Open Local ModelsMatthew Berman, Andriy Mulyar · Forward FutureThe Nomic AI co-founder explains how GPT4All emerged, why training data matters, and the tradeoffs between local and centralized AI.
40:25EN
No Priors: AI, Machine Learning, Tech, & Startups· 6. Apr. 2023Podcast · 2 StimmenA 6-Billion-Parameter Clone of ChatGPT’s Core TrickModeration: Sarah Guo, Elad GilDatabricks CTO Matei Zaharia tells Sarah Guo and Elad Gil how Dolly learned instruction-following on fifty times less data than GPT-3 — and why he would rather look facts up than bake them into a trillion-parameter network.
8:05ENMatthew Berman· 29. März 2023A 2023 Guide to Running the Original GPT4All Model LocallyMatthew Berman · Forward FutureMatthew Berman demonstrates an early CPU-based GPT4All release, its terminal installation and the resources available for fine-tuning.
1:21:20ENCognitive Revolution "How AI Changes Everything"· 9. März 2023PodcastFreeze the eyes and the brain. Train only the optic nerve.Moderation: Nathan LabenzJunnan Li and Dongxu Li tell Nathan Labenz how BLIP cleaned noisy web captions, and how BLIP-2 connected a frozen vision model to a frozen language model in under ten days on one A100.
53:37ENNo Priors: AI, Machine Learning, Tech, & Startups· 9. März 2023Podcast · 2 Stimmen30 models, 42 scenarios, 7 metrics: Percy Liang's public checkup for language modelsModeration: Sarah Guo, Elad GilA PubMed-only model once led the U.S. medical licensing exam — until a system roughly 200 times larger overtook it.
37:26ENNo Priors: AI, Machine Learning, Tech, & Startups· 23. Feb. 2023Podcast · 2 StimmenWithout Open Source, Machine Learning Would Be Decades BehindModeration: Sarah Guo, Elad GilHugging Face CEO Clem Delangue on the Tamagotchi pivot, why the company never hired a community manager, and why he does not believe one model will rule them all.Mehr laden24 von 198 angezeigt