test_dataset_Codellama-3-8B: benchmark results
An experimental Llama-3-8B-Instruct coding fine-tune released by rombodawg on Replete-AI/code-test-dataset. It is distinct from the later full Llama-3-8B-Instruct-Coder release. Provider: rombodawg. Access: Open.
Aggregate Rating 1415 ± 35, rank #5424 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 66.44 | Accuracy (%) (5-shot) | 87.3 |
| Open LLM Leaderboard v1 - GSM8K | 67.93 | Accuracy (%) (5-shot) | 84.7 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 52.43 | MC2 (%) (0-shot) | 55.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 60.58 | Normalized accuracy (%) (25-shot) | 48.4 |
| Open LLM Leaderboard v1 - WinoGrande | 75.06 | Accuracy (%) (5-shot) | 40.1 |
| Open LLM Leaderboard v1 - HellaSwag | 78.01 | Normalized accuracy (%) (10-shot) | 32.7 |
Interactive version: theaggregate.ai/model?slug=test-dataset-codellama-3-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.