© 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. 2994
    Prev
    Next
    INDEX
    Explanations

    preposition followed by noun/participle

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

    The neuron is detecting irregular past‐tense English verbs (e.g. “hit,” “came,” etc.).

    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
     namelijk
    -0.81
    fey
    -0.75
    نج
    -0.72
     out
    -0.69
    的时候
    -0.69
    ೨
    -0.69
    ה
    -0.69
    itia
    -0.69
     like
    -0.68
     enlighten
    -0.68
    POSITIVE LOGITS
     العديد
    0.93
    Byz
    0.88
     cabaña
    0.86
    Etymology
    0.85
    🤙
    0.85
     Kết
    0.84
    منی
    0.84
    coop
    0.83
    kB
    0.83
    orgue
    0.83
    Activations Density 0.067%

    No Known Activations