© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 28169
    Prev
    Next
    INDEX
    Explanations

    `ssp`

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

    The neuron is highly selective for the exact token “Type” (i.e. occurrences of the word “Type”).

    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
     hierbas
    -1.45
     stratég
    -1.41
     légèrement
    -1.40
    donde
    -1.39
    bisous
    -1.37
     autocollant
    -1.37
     botellas
    -1.34
    layın
    -1.32
     especias
    -1.32
     nyel
    -1.31
    POSITIVE LOGITS
    e
    1.84
    a
    1.70
    v
    1.63
    p
    1.62
    f
    1.61
    k
    1.61
    s
    1.58
    l
    1.58
    j
    1.57
    w
    1.52
    Activations Density 0.049%

    No Known Activations