© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 28013
    Prev
    Next
    INDEX
    Explanations

    release process

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

    This neuron fires most strongly on phrases describing the files that must be updated when incrementing or bumping the version number.

    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
    𝗺
    -1.11
    -
    -1.04
     hongos
    -1.01
    algebras
    -1.01
     trakcie
    -1.00
     or
    -0.99
    će
    -0.99
     familiar
    -0.98
     street
    -0.97
     dalam
    -0.97
    POSITIVE LOGITS
    ־ה
    1.34
    tbx
    1.18
    Público
    1.14
    меча
    1.13
    хождения
    1.11
    ᾿
    1.10
    њу
    1.07
     acceptez
    1.06
     MACHINE
    1.05
    dateien
    1.03
    Activations Density 0.164%

    No Known Activations