Technical report: Read paper · PDF · BibTeX · All models
Collection: Spot Snake models
Spot Snake MiniMind-JEV
cnfeichen/minimind-jev-v0.1 post-trained to pick moves in Spot's Biscuit Snake. It learns from the Spot Snake Decisions dataset: the game offers the safe moves and the model ranks them, copying a simple teacher that keeps the most open room.
Results
Teacher agreement: how often the model picks a move the teacher accepts. This is not game score, win rate or speed; those have not been measured.
| Check | Teacher agreement |
|---|---|
| Base model, validation | 56.30% (1,785 rows) |
| Best of 5 epochs, validation | 57.70% (1,785 rows) |
| Selected checkpoint, held-out test | 55.80% (1,708 rows) |
| Saved files reloaded, test subset | 56.25% (32 rows) |
Files
head.safetensors: trained ranking headencoder/: frozen BAAI/bge-small-zh-v1.5 encodermetrics.json,training.json,reload-check.json: what was measuredLICENSE,NOTICE,licenses/: licence texts
Use
hf download Lucasgitguy-co/spot-snake-minimind-jev-v0.1 --local-dir models/spot-snake-minimind-jev-v0.1
python scripts/serve.py --model models/spot-snake-minimind-jev-v0.1 # local game at http://127.0.0.1:8000
No GGUF: llama.cpp has no converter for the custom MiniMind ranking head.
The scripts, dataset generator and step-by-step instructions are in the
dataset repo (scripts/, INSTRUCTIONS.md).
The head is a ranker, not a text generator.
Limits
- Ranks only the moves the game already marks as safe; keep the game's collision checks.
- The teacher is a heuristic, not an optimal player. Matching it is the whole goal here.
- Confidence scores are not calibrated after training.
Licence
MIT: the MiniMind-JEV head, the BGE encoder and the new Spot Snake training are all MIT. See LICENSE, NOTICE and licenses/.
Base: cnfeichen/minimind-jev-v0.1 at fa26d3e20f.
{
"baseline_validation": {
"teacher_agreement": 0.5630252100840336,
"examples": 1785
},
"epochs": [
{
"teacher_agreement": 0.5607843137254902,
"examples": 1785
},
{
"teacher_agreement": 0.5568627450980392,
"examples": 1785
},
{
"teacher_agreement": 0.5647058823529412,
"examples": 1785
},
{
"teacher_agreement": 0.5770308123249299,
"examples": 1785
},
{
"teacher_agreement": 0.5703081232492997,
"examples": 1785
}
],
"test": {
"teacher_agreement": 0.5579625292740047,
"examples": 1708
}
}