© 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. 35286
    Prev
    Next
    INDEX
    Explanations

    questions and requests

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

    The neuron spikes on personal‐address words—especially first‐ and second‐person pronouns (I, you, them) and related function words used in direct, self‐referential or conversational statements.

    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
     comforting
    -0.80
     postes
    -0.77
     helps
    -0.74
     परि
    -0.74
     consolation
    -0.73
    Reliability
    -0.72
     confort
    -0.71
     importar
    -0.69
     encourages
    -0.69
     Preis
    -0.69
    POSITIVE LOGITS
     request
    1.05
     requests
    0.99
    Request
    0.96
    request
    0.95
    requests
    0.91
    REQUEST
    0.91
     submitted
    0.90
    setRequest
    0.84
     گفتم
    0.83
     requested
    0.82
    Activations Density 0.059%

    No Known Activations