© 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. 22-GEMMASCOPE-RES-131K
    7. 25286
    Prev
    Next
    INDEX
    Explanations

    choices and possibilities

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

    The neuron strongly activates on longer tokens (i.e. unusually long words or strings).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Ѕ
    -0.79
     jard
    -0.79
    planets
    -0.77
    ttemberg
    -0.76
    regated
    -0.75
    capa
    -0.75
     folle
    -0.74
     Lumpur
    -0.74
     集
    -0.73
    บอก
    -0.73
    POSITIVE LOGITS
     choices
    1.40
     possible
    1.40
     options
    1.38
     possibilities
    1.37
     Choose
    1.22
     choose
    1.21
     choice
    1.16
     chosen
    1.11
     combinations
    1.09
     optional
    1.07
    Activations Density 0.084%

    No Known Activations