Interconnects

109 Episoder

1. [Article Voiceover] Reverse engineering OpenAI's o1

   Publisert: 17.9.2024
2. Futures of the data foundry business model

   Publisert: 11.9.2024
3. A post-training approach to AI regulation with Model Specs

   Publisert: 10.9.2024
4. OpenAI's Strawberry, LM self-talk, inference scaling laws, and spending more on inference

   Publisert: 5.9.2024
5. OLMoE and the hidden simplicity in training better foundation models

   Publisert: 4.9.2024
6. On the current definitions of open-source AI and the state of the data commons

   Publisert: 28.8.2024
7. Nous Hermes 3 and exploiting underspecified evaluations

   Publisert: 16.8.2024
8. Interviewing Ross Taylor on LLM reasoning, Llama fine-tuning, Galactica, agents

   Publisert: 8.8.2024
9. A recipe for frontier model post-training

   Publisert: 7.8.2024
10. Interviewing Sebastian Raschka on the state of open LLMs, Llama 3.1, and AI education

    Publisert: 1.8.2024
11. GPT-4o-mini changed ChatBotArena

    Publisert: 31.7.2024
12. Llama 3.1 405b, Meta's AI strategy, and the new open frontier model ecosystem

    Publisert: 23.7.2024
13. SB 1047, AI regulation, and unlikely allies for open models

    Publisert: 17.7.2024
14. Switched to Claude 3.5

    Publisert: 3.7.2024
15. Interviewing Dean Ball on AI policy: CA SB 1047, upcoming AI disaster response, Llama 3 405B, Chinese open-source AI, and scaling laws

    Publisert: 27.6.2024
16. RLHF Roundup: Trying to get good at PPO, charting RLHF's impact, RewardBench retrospective, and a reward model competition

    Publisert: 26.6.2024
17. Frontiers in synthetic data

    Publisert: 21.6.2024
18. Text-to-video AI is already abundant

    Publisert: 18.6.2024
19. AI for the rest of us

    Publisert: 12.6.2024
20. A realistic path to robotic foundation models

    Publisert: 5.6.2024