Model Terminal
theo_em-optim-repro Qwen3-8B bad-medical Adam 2ep seed1
A fine-tuned Qwen3-8B checkpoint produced as part of a research reproduction study on emergent misalignment, investigating whether narrow harmful fine-tuning (here: 'bad-medical' examples, Adam optimizer, 2 epochs, seed 1) induces broader unsafe or deceptive behavior on unrelated prompts. This is a research artifact, not a production model. Source: supabase.
Value score
Not scored
Context
—
tokens
Max output
—
tokens
Price
—
Capability radar
Peer value bars
Identity
- Developer
- Misalignment-Empirics
- Openness
- —
- Modalities
- —
- Release
- —
- Knowledge cutoff
- —
- Deprecation
- —
- API docs
- —
Benchmarks
No benchmark scores yet.
Compare nearby
Pricing
- Input / 1M
- —
- Output / 1M
- —
- Speed
- —