© 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. 128634
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on first-person “I” statements and their associated self-referential commentary.

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

    i will

    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
    lasyon
    -1.34
    ּוֹ
    -1.27
     Polícia
    -1.17
    quisar
    -1.14
     auseinander
    -1.13
    再说
    -1.11
     Hochsch
    -1.07
    𝒙
    -1.06
     лука
    -1.05
     ſhe
    -1.05
    POSITIVE LOGITS
     I
    1.23
     گیاه
    1.20
     my
    1.19
    toJSON
    1.07
     Gegensatz
    1.05
    探索
    1.02
     lelaki
    1.00
    Ĝ
    0.99
    ῶν
    0.99
     even
    0.98
    Activations Density 0.026%

    No Known Activations