© 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. 34-GEMMASCOPE-RES-131K
    7. 77950
    Prev
    Next
    INDEX
    Explanations

    AmeriCorps and volunteer programs

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

    The neuron strongly activates on numeric tokens (e.g. “8,” “576,” “1552”) and all‐caps/acronym fragments (e.g. “COR,” “NOAA,” “PS”), i.e. on numbers and uppercase acronym pieces.

    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
     walnut
    -0.88
     Mimi
    -0.76
    Kubernetes
    -0.75
    nm
    -0.74
    standigheden
    -0.74
     Maxine
    -0.73
    ۱۲
    -0.73
    пче
    -0.71
     datter
    -0.71
     दोनों
    -0.70
    POSITIVE LOGITS
     Ameri
    2.84
    Amer
    2.03
     Amer
    1.97
     corps
    1.87
     Corps
    1.74
    Corps
    1.74
    corps
    1.55
     National
    1.43
     Corp
    1.41
     VISTA
    1.38
    Activations Density 0.022%

    No Known Activations