hugging-face-model-trainer

Supply Chain Compromise

Privilege Escalation

Data Exfiltration

Shadow Persistence

Resource Abuse

Prompt Injection

Privilege Escalation

Shadow Persistence

Guardrail Bypass

Supply Chain Compromise

Context Poisoning

Risky tool definition detected

- All dependencies in PEP 723 header - All build steps included - Clear error messages **Prevented failures:** Missing tokenizer libraries, build tool failures ## References **In this skill:** - `scripts/convert_to_gguf.py` - Complete, production-ready script **External:** - [llama.cpp Repository](https://github.com/ggerganov/llama.cpp) - [GGUF Specification](https://github.com/ggerganov/ggml/blob/master/docs/gguf.md) - [Ollama Documentation](https://ollama.ai) - [LM Studio](https://lmstudio.ai)

Data Exfiltration

Risky tool definition detected

Prompt Injection

Risky tool definition detected

TRL/Accelerate handles distribution automatically: ```python hf_jobs("uv", { "script": """ # Your training script here (same as single GPU) # No changes needed - Accelerate detects multiple GPUs """, "flavor": "a10g-largex2", # 2x A10G GPUs "timeout": "4h", "secrets": {"HF_TOKEN": "$HF_TOKEN"} }) ``` **Tips for multi-GPU:** - No code changes needed - Use `per_device_train_batch_size` (per GPU, not total) - Effective batch size = `per_device_train_batch_size` × `num_gpus` × `gradient_accumulatio

Supply Chain Compromise

Data Exfiltration

Shadow Persistence

Supply Chain Compromise

Resource Abuse

Data Exfiltration

Unauthorized Communication

Data Exfiltration

Unauthorized Communication

Resource Abuse

Supply Chain Compromise

Supply Chain Compromise

Data Exfiltration

Supply Chain Compromise

Data Exfiltration

Supply Chain Compromise

Risky tool definition detected

Tool passed security scan

Risky tool definition detected

### How to Apply **Known-good test inputs:** ```python # For training TEST_DATASET = "trl-lib/Capybara" # Small, well-formatted, widely used TEST_MODEL = "Qwen/Qwen2.5-0.5B" # Small, fast, reliable # For GGUF conversion TEST_ADAPTER = "evalstate/qwen-capybara-medium" # Known working model TEST_BASE = "Qwen/Qwen2.5-0.5B" # Compatible base ``` **Testing workflow:** 1.

Tool passed security scan