© 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. 34-GEMMASCOPE-RES-131K
    7. 1012
    Prev
    Next
    INDEX
    Explanations

    last names

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

    The neuron fires on long capitalized tokens—i.e. proper names (particularly surnames or other multi-letter named entities).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     surprisingly
    -1.18
    　　【
    -1.13
     only
    -1.05
     Surprisingly
    -1.03
     some
    -1.02
     unexpectedly
    -0.99
     strangely
    -0.97
     one
    -0.97
     now
    -0.95
     curiously
    -0.94
    POSITIVE LOGITS
     Jr
    1.48
     Obituary
    1.30
     III
    1.29
    Jr
    1.27
    othes
    1.27
     jr
    1.22
    oğlu
    1.16
    mäßigen
    1.15
    OfBirth
    1.12
     who
    1.09
    Activations Density 0.112%

    No Known Activations