Unverified85% confidenceFactExact time
The token-to-parameter ratios of Llama-2-7B and Gemma-2-9B are approximately 290 and 889, respectively, far exceeding Chinchilla's 20:1.
1
Sources
85%
Confidence
Long-term
Relevance
8/19/2026
First Seen
Sources
Rethinking Scaling Laws: Parameters Are Not the Only Answer
twitterjietang8/19/2026
Related Entities
Related Claims
VerifiedLlama was released by Meta, and from the Llama 1 to the Llama 3 series, parameter scales range from 7B to 405B65% similarUnverifiedQwen2.5-72B approaches the performance of Llama-3.1-405B across multiple benchmarks.64% similarUnverified在HumanEval、MBPP、GSM8K等基准测试上，Gemma 7B的表现显著优于同参数量级的LLaMA 2 7B和Mistral 7B61% similarUnverifiedLFM2.5-8B-A1B总参数量为8B，推理时仅激活1.5B参数60% similarUnverifiedMistral的7B模型在发布时超越了Llama 2 13B的表现58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/796929API
curl https://kongchang.com/api/v1/knowledge/claims/796929MCP
get_claim(id=796929)