© 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. 3538
    Prev
    Next
    INDEX
    Explanations

    Types of AFPs

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

    This neuron fires on abstract noun tokens (e.g. words naming concepts or qualities).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the phrase "set of" or similar variations indicating a collection or group.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    words signaling categorization or quantification—terms that denote sets, types, varieties, parts/percents, or example-introducing constructions.

    oai_token-act-pair · gpt-5Triggered 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
     সেই
    0.59
     সেটা
    0.59
     我
    0.58
     এই
    0.57
     dvije
    0.57
     そこ
    0.57
     nisam
    0.56
    我
    0.56
    两个
    0.55
     嗯
    0.55
    POSITIVE LOGITS
     billions
    0.53
     Britannica
    0.46
     railroads
    0.45
     laws
    0.45
     popularly
    0.44
     Bell
    0.43
     fluids
    0.41
     attacks
    0.41
     types
    0.40
     isotopes
    0.40
    Activations Density 0.001%

    No Known Activations