Current LLMs "undertrained by a factor of maybe 100-1000X or more" says OpenAI co-founder

Meta's Llama 3 was trained on a record amount of data, which could lead to a rethinking of the entire AI industry and produce better models.

The article Current LLMs "undertrained by a factor of maybe 100-1000X or more" says OpenAI co-founder appeared first on THE DECODER.

ai industry ai research ai training article artificial intelligence co-founder current data founder generative-ai industry llama llama 3 llms meta meta ai meta's llama 3 openai