© 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. 22-GEMMASCOPE-RES-131K
    7. 72850
    Prev
    Next
    INDEX
    Explanations

    official orders and memorandum

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

    The neuron is specialized in picking out standalone numeric tokens (metadata-like numbers) rather than normal words.

    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
    </strong>
    -2.36
    it
    -1.81
    0
    -1.78
    7
    -1.73
    the
    -1.69
    2
    -1.67
    6
    -1.65
    they
    -1.59
    5
    -1.58
     Many
    -1.55
    POSITIVE LOGITS
     Saltar
    1.95
     }}{\
    1.77
    嚙
    1.47
     nabíd
    1.45
    Ruju
    1.40
    好きです
    1.39
    mountains
    1.38
     débris
    1.38
    ציה
    1.34
    yordu
    1.34
    Activations Density 0.005%

    No Known Activations