Fine-tuning Foundations

When to fine-tune, data preparation, preference tuning, distillation, merging, evaluating and serving adapters.

1. 1▶ Up next14 min+50 XPWhen to fine-tune (and when not to)A decision framework across prompting, retrieval, fine-tuning and distillation: what each changes, cost and risk, the…
2. 2Pro18 min+50 XPData preparation and synthetic dataBuild a training set you can trust: sourcing and licences, formats and chat templates, cleaning, near-duplicate…
3. 3Pro17 min+50 XPPreference tuning and distillation: DPO, RLHF and small modelsLearn from comparisons rather than single answers: the preference data format, RLHF versus DPO with the DPO loss…
4. 4Pro16 min+50 XPMerging, evaluating and serving fine-tuned models and adaptersCombine models with weight averaging and task vectors (runnable), evaluate a fine-tune against baselines and for…