© 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. 22-GEMMASCOPE-RES-131K
    7. 34772
    Prev
    Next
    INDEX
    Explanations

    dollar amounts and thousands

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

    This neuron specializes in spotting numeric tokens (digits and multi‐digit numbers, often with commas).

    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
     also
    -1.73
     these
    -1.55
     really
    -1.53
     one
    -1.42
     remain
    -1.37
     actually
    -1.30
     understand
    -1.30
     will
    -1.27
    .......
    -1.27
     several
    -1.26
    POSITIVE LOGITS
     alemanes
    1.35
    还是很
    1.20
     sostu
    1.16
     vivimos
    1.16
     afectar
    1.14
    醣
    1.13
     constater
    1.13
     далее
    1.12
     breathtaking
    1.11
     europeos
    1.10
    Activations Density 0.031%

    No Known Activations