© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 31310
    Prev
    Next
    INDEX
    Explanations

    community words and descriptions

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

    The neuron strongly activates on the copular “is a” construction introducing an adjective or descriptor (e.g. “is a really special community”).

    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
     portato
    -0.85
     camicia
    -0.83
     parcheggio
    -0.82
     baggy
    -0.82
    lehem
    -0.81
    itisation
    -0.80
     kolejny
    -0.80
    ද
    -0.80
     coj
    -0.80
     computa
    -0.80
    POSITIVE LOGITS
     small
    1.05
     community
    1.01
     melting
    0.96
     farming
    0.93
    ımda
    0.92
     diverse
    0.91
     Christian
    0.90
     big
    0.83
    晏
    0.81
     tight
    0.81
    Activations Density 0.035%

    No Known Activations