© 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. 1433
    Prev
    Next
    INDEX
    Explanations

    What type

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

    The neuron isn’t responding to any lexical or semantic pattern in the words themselves but rather to their absolute position in the input—it fires strongly on tokens appearing at a particular offset in the document.

    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
     более
    0.81
     соотно
    0.74
    бычно
    0.73
     руководство
    0.68
    ือก
    0.68
     população
    0.67
    鶚
    0.67
    ровкой
    0.66
    ань
    0.66
    ("__
    0.65
    POSITIVE LOGITS
    ନ୍ଦ
    0.84
     spécial
    0.75
     Special
    0.74
    他们的
    0.74
     concerts
    0.74
     leur
    0.73
    OLOGIE
    0.72
     Dynamite
    0.70
    єю
    0.70
     chariot
    0.70
    Activations Density 0.000%

    No Known Activations