© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130121
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on first-person personal reflections or desires, i.e. “I …” statements describing what the writer used to do or ever wanted.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    what was wanted

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Ḧ
    -1.02
    aremos
    -0.98
     🤣
    -0.98
    潸
    -0.90
     😳
    -0.89
     Oggi
    -0.89
     😢
    -0.88
    EPI
    -0.88
     proph
    -0.88
     praktis
    -0.85
    POSITIVE LOGITS
     all
    1.16
    !
    1.11
     ideal
    1.01
     Geschenk
    0.98
     perfect
    0.96
    manchester
    0.90
     wanted
    0.90
     missing
    0.89
     Perfect
    0.88
     every
    0.87
    Activations Density 0.027%

    No Known Activations