© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 119563
    Prev
    Next
    INDEX
    Explanations

    burglary and arrest

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

    The neuron flags words and phrases that describe unauthorized or forced entry events—e.g. burglaries, break-ins, raids, and search warrants.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    {
    -1.45
    ();
    -1.23
    ،
    -1.23
    ifed
    -1.19
    点的
    -1.18
    无数
    -1.18
    "+"
    -1.18
     "/")
    -1.18
    》,
    -1.16
    tely
    -1.16
    POSITIVE LOGITS
    1
    1.67
    存档
    1.58
    Bereits
    1.48
    Муж
    1.42
    4
    1.40
    5
    1.38
     așa
    1.35
    ostante
    1.30
    kasten
    1.29
    confirmPassword
    1.25
    Activations Density 0.027%

    No Known Activations