© 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. 3120
    Prev
    Next
    INDEX
    Explanations

    generations

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

    This neuron isn’t picking out any semantic content at all but simply tracks where you are in the text (its activation grows roughly in step with the token’s absolute position).

    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
     pathophysiology
    0.76
     arbitrage
    0.76
    濯
    0.76
    ющие
    0.74
    जारत
    0.73
     hidrográf
    0.73
     CHCl
    0.73
     اقتصاد
    0.72
    нкү
    0.72
     juríd
    0.71
    POSITIVE LOGITS
    на
    1.01
    ن
    0.96
    ان
    0.95
    as
    0.93
    en
    0.92
    an
    0.91
    न
    0.89
    u
    0.89
    at
    0.89
    a
    0.85
    Activations Density 0.000%

    No Known Activations