Language Modeling on PG19 16K
8.01PerplexitySHAREDLLM
Evaluation results
| Method | Links | |
|---|---|---|
| SHAREDLLMBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 8.01 | |
| Activation BeaconBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 8.34 | |
| SHAREDLLMBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 9.02 | |
| StreamingLLMBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 9.25 | |
| Activation BeaconBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 9.41 | |
| StreamingLLMBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 9.63 | |
| LongAlpaca-16KBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 9.83 | |
| LongAlpaca-16KBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 10.39 |