Open runhani opened 4 months ago
https://arxiv.org/pdf/2306.15595
"max_position_embeddings": 131072, "model_type": "phi3_v", "original_max_position_embeddings": 4096,
https://arxiv.org/pdf/2402.13753
Extending Context Window of LLMs via Position Interpolation
https://arxiv.org/pdf/2306.15595
phi3-vision-128k
그래서 결론은?
LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens
https://arxiv.org/pdf/2402.13753