Fonderie · Atelier 01
Affinage refining
Training and fine-tuning open models on domain data, packaged as customised local models.
Affinage is the refining of molten metal: removing impurities until it reaches a specified composition.
État Draft. The run record, hyperspectral data handling, conversion of the HyperSIGMA base encoders to safetensors, three customisation methods and export with a model card work and are tested on synthetic data. No model trained on real data has been produced or released. This site is still an outline; detailed sections stay marked coming soon until they are written.
SommaireWhat this site will hold
Guide
From a domain dataset and a pinned base model to a customised local model, step by step.
- Installation
- A first run
- Run configuration
- Preparing domain data
- Training methods
- The run record
- Export as a local model
- The model card
- Licences and restricted data
Référence
Every public function, file format and schema, field by field.
- Python API
- Configuration keys
- Data manifest and split
- Run-record schema
- Export formats
Journal
What changed, when, and what was verified.
- Milestone notes
- Release notes
- Archived releases
PérimètreIntended scope
What Affinage is meant to do. The scope is written before the code, so the code can be held to it; none of it is implemented yet.
- Configuration-driven training and fine-tuning runs of open-weight models, full and parameter-efficient (for example LoRA), from a base model pinned to a commit hash.
- Data preparation with a manifest and checksums, and a recorded split that Essai can check for train–test overlap.
- A run record written next to every output.
- Export of the result as a customised local model: merged
safetensorsweights, an adapter, or a quantised file for a local runtime, each with a model card.
Not in scope
Hosted training services; weight inspection and surgery (Burin); benchmark evaluation (Essai).
JalonsFirst milestones
Run-record schema and writer, shared in form with Essai.
Coming soonA fine-tuning run on a tiny model and synthetic data that completes on CPU inside the test suite.
Coming soonExport with a generated model card.
Coming soon
La chaîneOne of three ateliers
-
Affinage refining
Training and fine-tuning open models on domain data, packaged as customised local models.
This site
Status: draft -
Burin engraving
Inspecting and editing the weights of open models.
Status: scaffold -
Essai assay
Reproducible evaluation of models on scientific tasks.
Status: scaffold
Every atelier follows the same rules for records, artefacts and safe loading: the Fonderie method ↗