Base model:finetune:huihui-ai/... Base model:huihui-ai/phi-4-abl... Endpoints compatible Lora Mergekit Peft Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/ngxson/LoRA-phi-4-abliterated
LoRA Phi 4 Abliterated Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
LoRA Phi 4 Abliterated Parameters and Internals
| LLM Name | LoRA Phi 4 Abliterated |
| Repository 🤗 | https://huggingface.co/ngxson/LoRA-phi-4-abliterated |
| Base Model(s) | |
| Required VRAM | 0.2 GB |
| Updated | 2025-09-07 |
| Maintainer | ngxson |
| Model Files | |
| Model Architecture | AutoModel |
| Is Biased | none |
| PEFT Type | LORA |
| LoRA Model | Yes |
| PEFT Target Modules | gate_up_proj|lm_head|o_proj|embed_tokens|qkv_proj|down_proj |
| LoRA Alpha | 32 |
| LoRA Dropout | 0 |
| R Param | 32 |
Best Alternatives to LoRA Phi 4 Abliterated
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Distil Longformer Base 4096 | 4K / 0.4 GB | 6 | 0 |
| Daedalus 1 | 1K / GB | 4 | 1 |
| Tiny Random Detr | 1K / 0.2 GB | 12 | 0 |
| Opengpt2 Pytorch Backward | 1K / 6 GB | 14 | 1 |
| Opengpt2 Pytorch Forward | 1K / 6 GB | 6 | 1 |
| Finsent Transformer | 0.5K / 0.4 GB | 2 | 1 |
| Bert Chinese L 12 H 768 A 12 | 0.5K / 0.4 GB | 4 | 1 |
| Simbert Chinese Tiny | 0.5K / 0 GB | 6 | 0 |
| Simbert Chinese Base | 0.5K / 0.4 GB | 6 | 0 |
| Bert Tiny | 0.5K / 0 GB | 10907259 | 126 |
Rank the LoRA Phi 4 Abliterated Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 51187 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124