TRL Training on Hugging Face Jobs
Train or fine-tune TRL language models on Hugging Face Jobs, including SFT, DPO, GRPO, and GGUF export.
When to Use
- The request matches the skill description: Train or fine-tune TRL language models on Hugging Face Jobs, including SFT, DPO, GRPO, and GGUF export.
- The task needs the implementation patterns, examples, validation checks, or edge cases listed in the topic map.
- The work would benefit from the complete guidance preserved in
references/full-guidance.md.
Core Workflow
- Confirm the request matches this skill's trigger, scope, and risk profile.
- Use the topic map to identify the relevant pattern, checklist, or example before writing detailed guidance or code.
- Load
references/full-guidance.md when implementation details, examples, anti-patterns, validation checks, or edge cases are needed.
- Apply only the relevant guidance instead of loading or repeating the entire reference by default.
- Verify the result against any validation checks, limitations, security notes, or platform constraints in the reference.
Topic Map
- Overview
- When to Use This Skill
- When to Use Unsloth
- Key Directives
- Local Script Execution
- Prerequisites Checklist
- ✅ Account & Authentication
- ✅ Dataset Requirements
- ⚠️ Critical Settings
- Asynchronous Job Guidelines
- Action Required
- Ground Rules
- After Submission
- Quick Start: Three Approaches
- Sequence Length Configuration
- Approach 1: UV Scripts (Recommended—Default Choice)
- Approach 2: TRL Maintained Scripts (Official Examples)
- Finding More UV Scripts on Hub
Reference Map
references/full-guidance.md preserves the complete original guidance, including examples and detailed edge cases.
Limitations
- Use this skill only when the task clearly matches the scope described above.
- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.
Progressive Loading
Keep this SKILL.md as the compact routing and workflow entrypoint. Load the reference file only when the user task requires the deeper implementation material.
1---2name: hugging-face-model-trainer3description: Train or fine-tune TRL language models on Hugging Face Jobs, including SFT, DPO, GRPO, and GGUF export.4license: MIT5---67# TRL Training on Hugging Face Jobs89Train or fine-tune TRL language models on Hugging Face Jobs, including SFT, DPO, GRPO, and GGUF export.1011## When to Use12- The request matches the skill description: Train or fine-tune TRL language models on Hugging Face Jobs, including SFT, DPO, GRPO, and GGUF export.13- The task needs the implementation patterns, examples, validation checks, or edge cases listed in the topic map.14- The work would benefit from the complete guidance preserved in `references/full-guidance.md`.1516## Core Workflow171. Confirm the request matches this skill's trigger, scope, and risk profile.182. Use the topic map to identify the relevant pattern, checklist, or example before writing detailed guidance or code.193. Load `references/full-guidance.md` when implementation details, examples, anti-patterns, validation checks, or edge cases are needed.204. Apply only the relevant guidance instead of loading or repeating the entire reference by default.215. Verify the result against any validation checks, limitations, security notes, or platform constraints in the reference.2223## Topic Map24- Overview25- When to Use This Skill26- When to Use Unsloth27- Key Directives28- Local Script Execution29- Prerequisites Checklist30- ✅ **Account & Authentication**31- ✅ **Dataset Requirements**32- ⚠️ **Critical Settings**33- Asynchronous Job Guidelines34- Action Required35- Ground Rules36- After Submission37- Quick Start: Three Approaches38- Sequence Length Configuration39- Approach 1: UV Scripts (Recommended—Default Choice)40- Approach 2: TRL Maintained Scripts (Official Examples)41- Finding More UV Scripts on Hub4243## Reference Map44- `references/full-guidance.md` preserves the complete original guidance, including examples and detailed edge cases.4546## Limitations47- Use this skill only when the task clearly matches the scope described above.48- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.49- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.5051## Progressive Loading52Keep this `SKILL.md` as the compact routing and workflow entrypoint. Load the reference file only when the user task requires the deeper implementation material.
newmindsgroup