© 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. 24-GEMMASCOPE-2-RES-16K
    7. 3369
    Prev
    Next
    INDEX
    Explanations

    indicating positive outcomes

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

    This neuron activates most strongly on low‐frequency or rare tokens—particularly proper names, technical terms, or unusual words.

    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_24_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
    ]}$
    0.79
    ر
    0.75
    i
    0.75
    lau
    0.75
     sitio
    0.71
     reactor
    0.70
     damen
    0.70
     amines
    0.69
     посу
    0.68
    ()}.
    0.68
    POSITIVE LOGITS
    ສ
    0.81
    がお
    0.80
     Pillow
    0.80
     Crav
    0.80
     Conflicts
    0.80
     oltre
    0.79
    احنا
    0.76
    купно
    0.75
     Dream
    0.74
     Oversight
    0.74
    Activations Density 0.000%

    No Known Activations