© 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. 22-GEMMASCOPE-RES-131K
    7. 54163
    Prev
    Next
    INDEX
    Explanations

    Czech Republic

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

    This neuron activates on the two-letter “grid position” labels (TL, TM, TR, ML, MM, MR, BL, BM, BR) used to index or name cells in a 3×3 layout.

    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
     conceito
    -0.79
     ballo
    -0.79
    }],
    -0.79
    pnea
    -0.79
     unique
    -0.77
     Zahn
    -0.77
    ーん
    -0.76
     supplements
    -0.76
     qual
    -0.75
    tigheter
    -0.75
    POSITIVE LOGITS
     mahdol
    0.86
    lores
    0.82
    abstra
    0.81
     SETTING
    0.81
    quese
    0.80
     undead
    0.78
    iev
    0.78
    🥋
    0.78
    dames
    0.78
    Kapcsolódó
    0.77
    Activations Density 0.016%

    No Known Activations