Model
GPT-2
OpenAI · closed weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| CNN/Daily Mail | 29.3 | rouge-1 | N/A | Independent | medium | d4mucfpksywv.cloudfront.net |
| DebateSum | 53.2 | rouge-l | N/A | Independent | medium | arxiv.org |
| language-modeling-recommendation | 48.8 | 1:1_accuracy | N/A | Independent | medium | arxiv.org |
| RAFT | 45.8 | avg | N/A | Independent | medium | arxiv.org |
| Text8 | 1.0 | bit_per_character_(bpc) | N/A | Independent | medium | d4mucfpksywv.cloudfront.net |
Math
MathMultimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| SIMMC2.0 | 19.2 | bleu | N/A | Independent | medium | d4mucfpksywv.cloudfront.net |
| VALSE | 60.1 | average_pairwise_accuracy | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Reddit Ideological Bias Dataset | 76.4 | weighted-f1_score | N/A | Independent | medium | sciencedirect.com |
| ThreatGram 101 | 66.2 | weighted-f1_score | N/A | Independent | medium | ieeexplore.ieee.org |
Reproduced by the community
Loading community runs…