メニュー
SkillsMP は agentculture/unsloth-cli から 6 件の skill を収集しています。skill を開くとソースと詳細を確認できます。
収集済み skill 6 件中 6 件を表示しています。
Explain how to use unsloth-cli — the agent-first CLI and its fine-tuning verbs. Covers the two faces of the tool (GPU-free introspection verbs: whoami / learn / explain / overview / doctor / cli; and the LoRA/QLoRA fine-tuning verbs: train / eval / export…
原文の言語: 英語
Drive a LoRA/QLoRA adapter training run end-to-end for this repo's fine-tuning verbs: validates the dataset and scope guard (GPU-free), resolves the plan via `sloth train --dry-run`, then (for a real run) runs the full adapter job (`sloth train`), evaluates…
原文の言語: 英語
Search the shared eidetic memory store and get back ranked, provenanced records. Drives `eidetic recall` with four search modes — exact (verbatim substring), approximate (vector/semantic), keyword (BM25 lexical), and hybrid (a weighted blend of…
原文の言語: 英語
Ingest records into the shared eidetic memory store so they can be recalled later. Drives `eidetic remember`: accepts one record as a JSON object, or a batch as NDJSON on stdin for bulk ingest. Upsert is idempotent by id (and dedups by content hash) —…
原文の言語: 英語
unsloth-cli's CI/CD lane, layered on `devex pr`. Delegates lint / open / read / reply / delta to devex; adds two extensions — `status` (SonarCloud quality gate + hotspots + unresolved-thread tally) and `await` (read --wait + status with non-zero exit on Sonar…
原文の言語: 英語
Cross-repo + mesh communication from unsloth-cli: file tracked GitHub issues on sibling repos, fetch issues from sibling repos to inline current state into briefs, and send live messages to Culture mesh channels. Use when the next step lives outside…
原文の言語: 英語