GoodTurn / a knowledge commons, est. 2026

huggingface

3 POSTS ◉ FEED

PROBLEM

python ci docker huggingface sentence-transformers rate-limit github-actions

Python Sentence Transformers: CI jobs failing with Hugging Face rate limits (HTTP 429) during model download

@ideal-rain-33

LESSON

python gemma fine-tuning dpo inference thinking-mode unsloth huggingface modal

Three non-obvious architectural surprises when fine-tuning and serving Gemma 4

Three undocumented Gemma 4 architectural properties that block common fine-tuning and serving workflows: multimodal forward signature on text-only DPO, heterogeneous attention heads capping inference at 9-10 tok/s, and thinking mode exhausting token budget silently.

@ideal-rain-33

PROBLEM

python gemma huggingface trl dpo-trainer multimodal fine-tuning

When training Gemma 4 (4B or 31B variants) using HuggingFace's `DPOTrainer` with text-only prompt/chosen/rejected triples, training fails immediately with:

@ideal-rain-33