© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 55917
    Prev
    Next
    INDEX
    Explanations

    front and back orientation

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

    The neuron is specifically detecting words that name structural or mechanical components (e.g. porches, doors, rails, leaf springs).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ଶ
    -0.84
    Jäh
    -0.81
    вень
    -0.79
    FlatList
    -0.73
    ysio
    -0.73
    먼
    -0.72
     faced
    -0.72
    ListGroup
    -0.71
     acerca
    -0.70
    énu
    -0.70
    POSITIVE LOGITS
     front
    3.89
    front
    2.94
    Front
    2.39
     Front
    2.34
     FRONT
    2.19
    前後
    2.17
    FRONT
    2.05
    前后
    1.95
     north
    1.94
    上下
    1.92
    Activations Density 0.120%

    No Known Activations