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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6821
    Prev
    Next
    INDEX
    Explanations

    internal vs external

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

    The neuron fires on language signaling changes to terms or policies—e.g. verbs and modals in “if we decide,” “we will change,” and similar update‐announcement phrasing.

    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
    film
    0.80
    lesias
    0.77
    }:
    0.75
    ﻥ
    0.75
    ⓡ,
    0.72
     adiabatically
    0.71
    separable
    0.70
    minute
    0.70
    methoxy
    0.70
    five
    0.69
    POSITIVE LOGITS
     tuve
    0.77
     히
    0.75
    ле
    0.70
    хо
    0.69
    ớm
    0.68
     BK
    0.66
     Коми
    0.66
     광
    0.66
     Tecn
    0.66
    га
    0.66
    Activations Density 0.001%

    No Known Activations