© 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. 65402
    Prev
    Next
    INDEX
    Explanations

    questions? comments? role

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

    The neuron is detecting numeric measurements and dosage/quantity instructions (e.g. amounts, percentages, times) in product-label or instruction text.

    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
     Alquiler
    -1.34
    Noticia
    -1.13
    </h1>
    -1.12
    occurs
    -1.10
     réputation
    -1.10
     acidente
    -1.10
    athed
    -1.08
     такъ
    -1.08
     Infrastruktur
    -1.07
     lakier
    -1.07
    POSITIVE LOGITS
     product
    1.33
    </em>
    1.24
     Перед
    1.24
    rekli
    1.24
    UserScript
    1.23
     while
    1.20
     scuro
    1.17
    stateMutability
    1.15
    õe
    1.12
     contains
    1.11
    Activations Density 0.009%

    No Known Activations