© 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. 10-GEMMASCOPE-RES-131K
    4. 102053
    Prev
    Next
    INDEX
    Explanations

    "this" followed by problem, issue, or goal

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

    the phrase "this" followed by a noun that represents a problem, gap, issue, or challenge being addressed in academic or technical writing.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on instances of “address” (and its inflections like “addresses”)—i.e. where a paper says it “addresses” a gap, problem, issue, or question.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     ส์
    -1.56
    îné
    -1.36
    quées
    -1.29
     Therefore
    -1.28
    গ
    -1.27
    塤
    -1.26
     dimanche
    -1.25
     Otro
    -1.24
     femminile
    -1.24
     Either
    -1.23
    POSITIVE LOGITS
     and
    2.38
     colocación
    1.57
     artesanía
    1.52
     ventilación
    1.51
     personalización
    1.48
     soldadura
    1.45
     opět
    1.42
     concentración
    1.42
     dlouho
    1.41
     espalda
    1.40
    Activations Density 0.061%

    No Known Activations