© 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. 31-GEMMASCOPE-2-RES-16K
    7. 9132
    Prev
    Next
    INDEX
    Explanations

    while affiliate

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

    The neuron detects phrases that delegate responsibility or choice—clauses like “leave it to …,” “it is up to …,” or “between … and …” that assign an action or decision to someone.

    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_31_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
     बोलते
    0.56
     createContext
    0.55
    ቾ
    0.55
    이가
    0.53
    LOP
    0.53
     raíz
    0.52
     Среди
    0.52
     வேகமாக
    0.52
     Muitos
    0.51
    рованное
    0.51
    POSITIVE LOGITS
     leave
    1.31
     Leave
    1.31
    Leave
    1.24
    leave
    1.23
     entrust
    1.20
    交给
    1.18
    留
    1.17
     up
    1.16
     onus
    1.12
     responsibility
    1.11
    Activations Density 0.390%

    No Known Activations