© 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. 73531
    Prev
    Next
    INDEX
    Explanations

    more comparisons

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

    The neuron is triggered by second-person pronouns (especially “you”) and closely related personal pronouns and verbs in direct address.

    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
    ναι
    -0.94
     недорого
    -0.93
    ñadir
    -0.91
     RSSSF
    -0.88
    ַּ
    -0.87
     ;)
    -0.84
    ább
    -0.84
    给你
    -0.83
    谩
    -0.83
    verket
    -0.82
    POSITIVE LOGITS
     I
    1.10
    Whereas
    1.08
    whereas
    1.03
     whereas
    1.02
    compared
    1.01
    Compared
    0.99
     sedangkan
    0.99
     my
    0.96
     better
    0.94
    apparently
    0.94
    Activations Density 0.083%

    No Known Activations