Inspect the measurement.

Qwen/Qwen2.5-0.5B-Instruct · HellaSwag (full validation, character-normalized accuracy) · 52.4% accuracy_char_normalized

Full benchmark split · 10042 scored examples. Full pinned split; primary character-normalized accuracy: argmax summed conditional answer-token log likelihood divided by Python len(choice), excluding the leading delimiter; raw accuracy is a separate diagnostic. Explicit pinned prompt schema, no chat/BOS. Left context truncated only if prompt plus answer exceeds 1024 tokens. Generic text diagnostic does not validate specialist declared tasks.

Run ID / completed

b6d519b9ceea361153040727
17 Oct 2026, 15:23 UTC

Model revision

7ae557604adf67be50417f59c2c2f167def9a775

Comparison scope

9ba01dfd8954831947fc9d9e821f5c27a78509f522b87d05d07b1c4f172ae0e6

Evaluator SHA-256

8ba81403e69d93b8121bd05e0ea4f9c55e0dc5c721449f50adfda136d7d976f1

Observed resources

328s elapsed · 3.31 GiB peak torch allocation
Dollar cost: not measured

Download evidence

result.json samples.jsonl environment.json evaluate.py manifest.json

Inspect individual examples

Read the recorded prediction, reference and outcome for each example.

Scored examples: 10,042
Passed: 5,267
Failed: 4,775
Metric: accuracy_char_normalized

Examples are downloaded only when requested. Search covers all loaded records; 25 examples are displayed per page. Download all examples · JSONL

Example evidence has not been loaded.