Language Modeling on ProofPile (16K)
3.24PerplexitySHAREDLLM
Evaluation results
| Method | Links | |
|---|---|---|
| SHAREDLLMBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 3.24 | |
| Activation BeaconBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 3.34 | |
| LongAlpaca-16KBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 3.34 | |
| LongAlpaca-16KBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 3.37 | |
| SHAREDLLMBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 3.38 | |
| StreamingLLMBase Model=LLaMA-2, Supervised fine-tuning=true2026.03 | 3.51 | |
| Activation BeaconBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 3.64 | |
| StreamingLLMBase Model=Mistral-7B, Supervised fine-tuning=true2026.03 | 4.19 |