domains / families / research / colm-experiments

colm-experiments

separate

Use when designing or auditing the empirical core of a COLM paper — contamination analysis for evaluation data, fair baselines under matched prompting and compute, pinned model versions and decoding parameters, uncertainty over runs and samples, scaling coverage, and honest reporting of API-model comparisons.

When to use

• designing empirical core of COLM paper
• auditing empirical core of COLM paper
• contamination analysis for evaluation data
• fair baselines under matched prompting and compute
• pinned model versions and decoding parameters
• uncertainty over runs and samples
• scaling coverage
• honest reporting of API-model comparisons

Skill page

Source