71 skills · sorted by relevance
Write and debug code for further training large language models with TRL. Match training methods to dataset formats, configure supervised or preference-based training, and set up reward functions through Python or the command line.
Write and debug code for further training large language models with TRL. Match training methods to dataset formats, configure supervised or preference-based training, and set up reward functions through Python or the command line.
Add or review research paper entries in TRL’s paper index. Organize each paper’s contribution, its connection to the implementation, and a Python configuration example using training settings supported by the paper.
Guide language model training with TRL using example answers, preferred response pairs, or reward scores. The skill covers data preparation, method selection, reward model training, evaluation, and adjustments for memory limits or unstable training.
Guides language model training with GRPO in TRL, comparing several responses and rewarding better results. Covers dataset preparation, scoring rules for correctness and format, training configuration, monitoring, and saving models for later use.
Guide language model training with TRL, from learning to follow instructions to matching human preferences. Covers method selection, training data, reward models that score responses, evaluation, and adjustments for memory limits or unstable training.
Fine-tune language models with TRL using GRPO, a method that compares scored responses to the same prompt. Covers training data, scoring functions, GPU-conscious settings, and monitoring for tasks with measurable correctness or required output formats.
Guide language model training with TRL, from learning to follow instructions to matching human preferences. Covers method selection, training data, reward models that score responses, evaluation, and adjustments for memory limits or unstable training.
Customize Claude Code keyboard shortcuts while preserving existing settings. Change submission keys, add chords—shortcuts entered as a sequence of keystrokes—and remove default bindings, with checks for invalid settings and terminal or operating system conflicts.
Train or fine-tune language and vision models on Hugging Face Jobs without a local GPU. Covers dataset checks, hardware and cost planning, training monitoring, Hub storage, and GGUF conversion for local use.
Execute keyboard shortcuts. When you need to copy/paste (Ctrl+C/V), save files (Ctrl+S), close windows (Alt+F4), undo/redo (Ctrl+Z/Y), or select all (Ctrl+A).
The English summary is being prepared. Showing the original description.
Reference for the TRL (Transformer Reinforcement Learning) library codebase. Use proactively before reading or editing any file under `trl/` so you have the intended contracts and invariants in mind, not just what the current code says. Covers trainer hierarchy (SFT, DPO, GRPO, KTO), shared utility functions (selective_log_softmax, decode_and_strip_padding, padding helpers), configuration system, model wrappers, and how data flows through any TRL trainer.
The English summary is being prepared. Showing the original description.
Prepare GRPO reinforcement learning configurations for tasks with programmatically checkable answers, such as math, code, and structured output. Design and inspect reward functions, then select training variants when runs show instability or exploit flaws in scoring.
⧉ Copy a message to the system clipboard cleanly formatted — no blockquote pipes, no terminal bars, no Claude commentary. Use whenever you draft a message the user will paste (Slack, Teams, PR comment, email) to add a clickable "⧉ ctrl + click" copy link under it, and when the user says "/ctrl-clip", "copy that", "put it on the clipboard" (in any language). Optional argument picks the target format: slack (default), md, plain.
The English summary is being prepared. Showing the original description.
Ajuste fino de LLMs usando reinforcement learning com TRL - SFT para ajuste de instruções, DPO para alinhamento de preferências, PPO/GRPO para otimização de recompensas e treinamento de modelo de recompensas. Use quando precisar de RLHF, alinhar modelo com preferências ou treinar a partir de feedback humano. Funciona com HuggingFace Transformers.
The English summary is being prepared. Showing the original description.
Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.
The English summary is being prepared. Showing the original description.
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
The English summary is being prepared. Showing the original description.
Explore training workflows that reduce harmful model responses using a written set of principles. The skill covers self-critique, revised training examples, and RLAIF, reinforcement learning from AI feedback, with implementation examples and hardware guidance.
Explore training workflows that reduce harmful model responses using a written set of principles. The skill covers self-critique, revised training examples, and RLAIF, reinforcement learning from AI feedback, with implementation examples and hardware guidance.
Convert existing Hugging Face Transformers Trainer or TRL SFTTrainer training code into an NVFLARE federated job using flare.patch(trainer), local validation, and job export; use when the user names Hugging Face or preliminary source inspection identifies one Hugging Face owner, and not for manual PyTorch loops, Lightning, inference-only pipelines, deployment, or experiment workflows.
The English summary is being prepared. Showing the original description.
Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.
The English summary is being prepared. Showing the original description.
Use when changing REPL clear/reset/resume/surface behavior (onClearTerminal, transcriptSeq remount, Ink Static paths, Ctrl+O/Ctrl+E transitions). Enforces single-owner reset transaction and prevents black-screen/flicker/duplicate-row regressions.
The English summary is being prepared. Showing the original description.
Expert knowledge of Folketrygdloven (National Insurance Act) processing in melosys-api. Use when: (1) Understanding FTRL bestemmelser (§2-5 to §2-13), (2) Debugging yrkesaktiv/ikke-yrkesaktiv/pensjonist processing flows, (3) Understanding trygdedekning options and combinations, (4) Investigating vilkår for each bestemmelse, (5) Understanding medlemskapsperiode generation logic, (6) Debugging frivillig vs pliktig membership determination.
The English summary is being prepared. Showing the original description.
TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.
The English summary is being prepared. Showing the original description.