© 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. 7813
    Prev
    Next
    INDEX
    Explanations

    hypothetical questions and assumptions

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

    The neuron detects adverbial or discourse‐level qualifiers expressing frequency, likelihood, or speaker stance (e.g. “frequently,” “assuming,” “most likely”).

    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
    ствовали
    0.55
    醚
    0.55
    Listed
    0.55
     Listed
    0.55
     occasioned
    0.55
     listed
    0.54
    GRADE
    0.54
    ኹ
    0.54
     Within
    0.52
     alleviated
    0.52
    POSITIVE LOGITS
     ricevuto
    0.57
     unequivocal
    0.55
    隣
    0.54
     brown
    0.53
    يال
    0.53
     मानस
    0.53
     gh
    0.52
     volle
    0.52
     cardiac
    0.51
     estiver
    0.51
    Activations Density 0.000%

    No Known Activations