© 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. 10-GEMMASCOPE-RES-131K
    7. 73540
    Prev
    Next
    INDEX
    Explanations

    greeting followed by name

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

    The neuron strongly activates on common greeting or salutation words (e.g. “Hello,” “Hi,” “Howdy,” “Здравствуйте”) at the start of a message.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     sorprend
    -1.82
     zimowe
    -1.80
    
    -1.77
     policiers
    -1.70
     this
    -1.70
     it
    -1.69
    نگار
    -1.63
     esconde
    -1.62
     représentants
    -1.53
     étapes
    -1.52
    POSITIVE LOGITS
     Vielzahl
    2.22
    躄
    1.84
     و
    1.73
    otheses
    1.70
    ⁛
    1.70
     juſ
    1.68
    время
    1.66
     vorhandenen
    1.66
    jenigen
    1.60
    ukone
    1.59
    Activations Density 0.079%

    No Known Activations