Audit a Gemma Tuner plan phase against its acceptance criteria, tests, artifacts, leakage controls, hardware claims, and stop conditions.
原文语言:英语
菜单
SkillsMP 已收集 mattmireles/gemma-tuner-multimodal 中的 12 个 Skill。打开任一 Skill 可查看来源和详情。
已展示 12 / 12 个已收集 Skill。
Audit a Gemma Tuner plan phase against its acceptance criteria, tests, artifacts, leakage controls, hardware claims, and stop conditions.
原文语言:英语
Execute an approved Gemma Tuner implementation or experiment plan phase by phase with frozen gates and recorded evidence.
原文语言:英语
Record Gemma Tuner investigations, experiment evidence, failures, and operational learnings in the repo's existing notes structure.
原文语言:英语
Edit Gemma Tuner Markdown while preserving links, runnable commands, tables, heading structure, and honest model/data/hardware claims.
原文语言:英语
Audit Gemma Multimodal Fine-Tuner for training correctness, dataset leakage, MPS/device behavior, checkpoint/export integrity, evaluation validity, privacy, and CLI/wizard/visualizer regressions.
原文语言:英语
Debug Gemma Multimodal Fine-Tuner failures across Python environments, Hugging Face models, PEFT/LoRA, datasets, audio/image processors, MPS, wizard subprocesses, visualizer state, evaluation, or export.
原文语言:英语
Write or update Gemma Tuner documentation for CLI, configuration, datasets, training, evaluation, export, MPS behavior, wizard, and visualizer changes.
原文语言:英语
Carry a Gemma Tuner plan through implementation, empirical verification, adversarial audit, and honest closeout. Use only when explicitly invoked. A preregistered KILL is a valid completed result.
原文语言:英语
Create a scoped Gemma Tuner commit after tests pass. Use only when the user explicitly asks to commit.
原文语言:英语
Push a tested Gemma Tuner branch after confirming commit scope and remote state. Use only when explicitly requested.
原文语言:英语
Debug failures that cross Gemma Tuner and external Hugging Face, Transformers, PEFT, PyTorch MPS, dataset, or model-card boundaries.
原文语言:英语
Plan external research for Gemma, Hugging Face, PEFT, PyTorch MPS, multimodal processors, or training behavior when current authoritative knowledge is required.
原文语言:英语