fine-tuning-with-trl
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.
- Score
- Unranked
- Evidence
- No signals yet
- Last commit
- Not known
- Listed
Install
armory install fine-tuning-with-trl --cli claudewrites the skill into.claude/skills/fine-tuning-with-trl/SKILL.mdListed as compatible
# fetches the source and writes it to:
.claude/skills/fine-tuning-with-trl/SKILL.mdNeeds the armory CLI · not on npm yet, build it from cli/ in the repository
What it is
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.
When to use it
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.
How to install / invoke
# Copy the skill into your .claude/skills/ directory
curl -sL https://raw.githubusercontent.com/davila7/claude-code-templates/main/cli-tool/components/skills/ai-research/post-training-trl-fine-tuning/SKILL.md -o .claude/skills/post-training-trl-fine-tuning/SKILL.md
Notes
Extracted from davila7/claude-code-templates, post-training-trl-fine-tuning category.