LinMix - OLMo2 1B 288 Domains - hellaswag BPB: leaderboard
LinMix compares separately trained, publicly released language models built from scratch with different training-data mixtures. Each result belongs to its exact trained checkpoint; lower bits per byte is better. Bits per byte (lower is better).
Metric: Bits per byte (lower is better). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | LinMix olmo2-1B-288domains-vertex-linear-hellaswag | 0.76 | |
| 2 | LinMix olmo2-1B-288domains-vertex-linearsqrt-hellaswag | 0.76 | |
| 3 | LinMix olmo2-1B-288domains-vertex-linearscaled-hellaswag | 0.76 | |
| 4 | LinMix olmo2-1B-288domains-vertex-linearlog-hellaswag | 0.76 | |
| 5 | LinMix olmo2-1B-288domains-dirichlet-linearscaled-hellaswag | 0.76 | |
| 6 | LinMix olmo2-1B-288domains-dirichlet-linear-hellaswag | 0.76 | |
| 7 | LinMix olmo2-1B-288domains-dirichlet-exponential-hellaswag | 0.78 | |
| 8 | LinMix olmo2-1B-288domains-dirichlet-lightgbm-hellaswag | 0.78 | |
| 9 | LinMix olmo2-1B-288domains-baseline | 0.79 |
Interactive version: theaggregate.ai/benchmark?slug=linmix-olmo2-1b-288-domains-hellaswag-bpb · How It Works · Data refreshed daily, snapshot 2026-10-29.