GPT-2-Large-40k-steps: benchmark results
Provider: OpenAI. Access: API.
Aggregate Rating 1082 ± 42, rank #10786 of 10870 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 47.84 | MC2 (%) (0-shot) | 38.9 |
| Open LLM Leaderboard v1 - GSM8K | 0 | Accuracy (%) (5-shot) | 4.6 |
| Open LLM Leaderboard v1 - HellaSwag | 26.55 | Normalized accuracy (%) (10-shot) | 3.4 |
| Open LLM Leaderboard v1 - MMLU | 24.15 | Accuracy (%) (5-shot) | 3.1 |
| Open LLM Leaderboard v1 - WinoGrande | 49.88 | Accuracy (%) (5-shot) | 2.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 22.87 | Normalized accuracy (%) (25-shot) | 2 |
Interactive version: theaggregate.ai/model?slug=gpt-2-large-40k-steps · How It Works · Data refreshed daily, snapshot 2026-11-12.