Tensor-Slayer: LLM model editing directly on the weights, without fine-tuning
Tensor-Slayer is an MIT-licensed tool for exploring and editing a language model at the tensor level. An agent analyses tensors (statistics, distributions, patterns), recommends targeted edits, and applies them to the weights file with before-and-after previews, safety checks and automatic backups, with no fine-tuning or dataset. It started a line of work that runs through rank-one fact edits and layer-stacked merges.
Fine-tuning changes a model by showing it data. Tensor-Slayer asks a different question: how far can you get by reading the weights and editing them directly?
What the tool does
- Explore a model's structure, with a focus on attention, MLP and embedding tensors.
- Analyse individual tensors: minimum, maximum, mean, standard deviation, sparsity, percentile distributions, pattern detection and simulated operations.
- Patch with guard rails: choose an operation (scale, add, normalise), a target and a value, preview the before and after statistics, pass safety checks, and keep an automatic backup.
- Adapt edits to position: layer-aware modifications based on where a tensor sits and what role it plays.
- Ask an agent for recommendations aimed at a capability such as maths or reasoning.
The repository includes the patches and the evaluation results it reports on Qwen 0.6B (MMLU and HumanEval). Those are the repository's own measurements; treat them as a starting point, not a benchmark claim.
The line of work
- Editing quantised GGUF weights in a UI; a C runtime for running Gemma on a laptop GPU
- Frankenmerge: layer-stacked Qwen and GLM merges, healed with SFT and RL
- Rank-one fact edits on Qwen3 (Qwen-ROME, public)
- Surgical Theatre: a zero-copy context manager for temporary weight edits during RL (public)
- Tensor-Slayer 0.2: DSPy generates the patches, nothing hard-coded
- Autonomous pruning: a DSPy agent chooses the structured-pruning strategy, processing layer by layer for models larger than RAM
- Inferenceless interpretability: point at a local checkpoint, explore its tensors and discuss the architecture with an assisting LLM
- Statistical tensor analysis and merging across incompatible architectures
- Tensor-Slayer released: an agent proposes targeted tensor operations
- Hex-patching QwQ-32B: a tensor explorer and forward-pass tracer
Facts
| Item | Detail |
|---|---|
| Repository | areu01or00/Tensor-Slayer, MIT licence |
| Format | Works on safetensors files directly |
| Related public work | Qwen-ROME, tensorslayereval |
What you can buy
Environments where an agent has to change a model's behaviour under constraints: edit or merge weights, keep a frozen evaluation from regressing, and stay within a compute budget. Graders come from the evaluation set and the constraints.
FAQ
Can you improve a model without fine-tuning?
Sometimes, in narrow ways. Targeted edits to specific tensors, rank-one fact edits and merges can change behaviour without a training run. Whether an edit helps has to be measured on a held-out evaluation, every time.
What is ROME model editing?
Rank-One Model Editing changes a single stored fact by applying a rank-one update to one MLP layer's weights. Qwen-ROME applies it to Qwen3 models.
Is Tensor-Slayer open source?
Yes, under the MIT licence.
License it, or have me build an environment in the same style.