© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6675
    Prev
    Next
    INDEX
    Explanations

    dogs, puppies, or hounds

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on passages describing dogs—especially when naming breeds, litters, or their roles (e.g. “livestock guardian dogs,” puppies, healthy pups, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mèo
    0.71
     transcoding
    0.65
     diken
    0.64
     veterinary
    0.64
     vertebrate
    0.63
     बांस
    0.63
     रूई
    0.62
    캣
    0.61
     veterinarians
    0.60
     prebiotic
    0.60
    POSITIVE LOGITS
     puppies
    1.19
     bred
    1.14
     breed
    1.12
     puppy
    1.11
    breed
    1.08
     breeding
    1.06
     dogs
    1.05
    Dogs
    1.05
     breeders
    1.04
    Breed
    1.02
    Activations Density 0.243%

    No Known Activations