SmolLM2-360M-Instruct
HuggingFaceTB/SmolLM2-360M-Instruct
0.4BSmolLM3 Heretic artifacts2 measured conversions3K downloads
Best version97.2Pareto
saidutta69/SmolLM2-360M-Instruct-hereticby saidutta69 · Heretic v1.4.0 · GGUF, SAFETENSORS
- KL divergence
- 0.015
- Refusals
- 4/100
- Refusal rate
- 4%
- Downloads
- 571
Evidence from the model card
| KL divergence | 0.0154 | | Refusals | 4/100 |
Metrics published by the author. Why this ranks first.
All measured versions
Higher score is better. One row per ablation attempt; artifacts of the same attempt are grouped under it.
| # | Author | Params | KL | Refusals | Score | Evidence | Downloads |
|---|---|---|---|---|---|---|---|
| 1 | saidutta69 | 0.4B | 0.015 | 4/1004% | 97.2 | BestParetoGGUFSAFETENSORS | 571 |
| 2 | isotnek | 0.4B | 0.019 | 5/1005% | 96.5 | SAFETENSORS | 584 |
All artifacts (3)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| SmolLM2-360M-Instruct-Heretic-GGUF mradermacher | quant | GGUFIQ4_XS | none | 1.8K | 2026-02-21 |
| SmolLM2-360M-Instruct-heretic isotnek | heretic | ONNXSAFETENSORS | published (weak match)KL 0.019 | 584 | 2025-12-15 |
| SmolLM2-360M-Instruct-heretic saidutta69 | heretic | GGUFSAFETENSORSF16 | published (weak match)KL 0.015 | 571 | 2026-10-10 |
1 artifact inherit metrics from their parent conversion or publish none.