Protocol
Language Model Adaptation and Scaling
Adapt a small language model to a documented domain and audit the evaluation for contamination. Compare capability gains with resource use and regression on an unrelated holdout.
50–85 active hours60–90 min sessionsAdvanced
Mastery contract
Specialization · Standard 1.0 · Adapt a small language model to a documented domain and audit the evaluation for contamination. Compare capability gains with resource use and regression on an unrelated holdout.
Loading your learning records…
Reasoning — not yet passed
Compute parameter counts and explain the low-rank constraint without claiming all useful updates must be low rank.
Reliable implementation — not yet passed
Merged and unmerged outputs agree within a stated tolerance and only intended parameters receive gradients.
Reproducible experiment — not yet passed
Every rank and baseline is reported with tokenizer, data provenance, tokens processed and memory usage.
Defense and handoff — not yet passed
Defend adaptation and compute-allocation choices, explicitly limiting conclusions to the tested scale and domain.
All four criteria must pass. Activity completion does not satisfy them.
An independent reviewer reruns your work and varies something you did not rehearse: a fresh input, a different working directory, or a declared edge case.
Pilot decisions are provisional, not external credentials. For an appeal or sensitive artifact, contact your designated pilot operator with the submission ID. Export your assessment history.
What you will do
6 total
Derive the LoRA weight update, its parameter count and a forward FLOP estimate; distinguish a scaling-law fit from an established causal law.
Implement LoRA on selected projections of a small Transformer; verify zero-update initialization and merged/unmerged prediction equivalence.
Compare frozen, LoRA and full fine-tuning under fixed data and a documented compute budget; vary 3 ranks and report memory, held-out loss and overfit behavior.
Debug training-set documents leaked into evaluation and an adaptation comparison with unequal token budgets.
Write a 500–900 word technical report linking the reasoning, code, measurements, and limitations; include the project decision below.
Assess all 4 mastery criteria against saved artifacts, request an independent review, and repeat each failed criterion on a fresh example.
Essentials
A reading session alone never satisfies a mastery criterion.
Store source references, assumptions, and artifact paths beside every result.
Keep an untouched check case that differs from the worked example.
Record all attempts, including failures and results that contradict your prediction.
Use the stated comparison conditions; document every deviation before drawing a conclusion.
Protocol guardrails
Do
- +State the expected result before running the comparison.
- +Keep one minimal reproducible failing case when debugging.
- +Record environment versions and the exact command used.
- +Compare explanations with saved intermediate values.
- +Ask a reviewer to challenge the weakest assumption.
Don't
- ×Do not copy a worked solution and present it as an independent implementation.
- ×Do not tune against held-out evaluation outcomes.
- ×Do not report only the best seed or discard inconvenient runs.
- ×Do not equate elapsed hours or a completed run with a passed assessment.
- ×Do not conceal reduced-scale experiments behind claims about the original full-scale result.
Protocol authorship
Written by NuthinButta as instructional design. The teaching sources are the official references linked in each milestone; the exercises, workload and pass thresholds are ours, not their authors'.