Surprising Model Performance
Tatsu shares how their model surprised them by handling diverse tasks beyond expectations with minimal data, highlighting the importance of a strong base model. The shift to the llama model marked a significant improvement in capabilities, showcasing the impact of fine-tuning on performance.In this clip
From this podcast
Unsupervised Learning
Ep 11: Stanford Professor Tatsu Hashimoto on AI Biases and Improving LLM Performance
Related Questions
What are the costs of AI models?
Why are deep learning models so expensive that only big players in tech can afford to develop them, as discussed in the episode Ben Green: "Tech for Social Good" Needs to Do More and the clip The Power of Big Data?
What are the costs of AI models mentioned in the episode Using Large Language Models at AngelList // Thibaut Labarre // MLOps Podcast #171 and the clip Empowering with Language Models?