GPT-2-Large-20k-steps: benchmark results
Provider: OpenAI. Access: API.
Aggregate Rating 1078 ± 43, rank #10811 of 10870 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 50.99 | MC2 (%) (0-shot) | 50.3 |
| Open LLM Leaderboard v1 - GSM8K | 0 | Accuracy (%) (5-shot) | 4.6 |
| Open LLM Leaderboard v1 - MMLU | 23.84 | Accuracy (%) (5-shot) | 2.5 |
| Open LLM Leaderboard v1 - ARC Challenge | 22.78 | Normalized accuracy (%) (25-shot) | 1.9 |
| Open LLM Leaderboard v1 - HellaSwag | 25.81 | Normalized accuracy (%) (10-shot) | 1.8 |
| Open LLM Leaderboard v1 - WinoGrande | 48.46 | Accuracy (%) (5-shot) | 0.5 |
Interactive version: theaggregate.ai/model?slug=gpt-2-large-20k-steps · How It Works · Data refreshed daily, snapshot 2026-11-12.