Language Modeling on PG19 4K
8.68PerplexitySHAREDLLM
Evaluation results
| Method | Links | |
|---|---|---|
| SHAREDLLMBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 8.68 | |
| SHAREDLLMBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 8.97 | |
| StreamingLLMBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 9.21 | |
| Activation BeaconBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 9.21 | |
| Activation BeaconBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 9.35 | |
| StreamingLLMBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 9.58 | |
| LongAlpaca-16KBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 9.96 | |
| LongAlpaca-16KBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 10.21 |