© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9031
    Prev
    Next
    INDEX
    Explanations

    what is needed

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

    This neuron responds strongly to rare or uncommon tokens (i.e. it flags low‐frequency 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
     carcinogenic
    0.86
     expansive
    0.82
    帏
    0.82
     tritium
    0.80
    烜
    0.80
     fullest
    0.78
     mountainous
    0.78
     affecting
    0.78
     Spacious
    0.78
     Radiant
    0.77
    POSITIVE LOGITS
    isme
    0.77
    τα
    0.77
    te
    0.76
    Ped
    0.76
    ñ
    0.73
    crm
    0.71
     fortæ
    0.70
    sson
    0.67
    aa
    0.67
     gøre
    0.67
    Activations Density 0.000%

    No Known Activations