© 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. 65360
    Prev
    Next
    INDEX
    Explanations

    environment, website, cake, least, medium, user

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

    This neuron activates on nouns that denote a setting, platform, or tool (e.g. “environment,” “website,” “solution”).

    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
     indeks
    -1.13
    Ќ
    -1.11
    ariConfig
    -1.11
     различи
    -1.07
    뭄
    -1.06
     eksklu
    -1.04
    mappedBy
    -1.04
     visuel
    -1.02
    kmale
    -1.00
    玎
    -0.99
    POSITIVE LOGITS
     agregó
    1.08
     añadió
    1.05
     manifestó
    1.00
    问问
    0.98
     cuidar
    0.96
    🪚
    0.91
     wunderbar
    0.90
     zusammen
    0.89
    ]
    0.88
    珩
    0.87
    Activations Density 0.089%

    No Known Activations