© 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. 92471
    Prev
    Next
    INDEX
    Explanations

    programming and non-english characters

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

    The neuron fires on the “information‐heavy” or unusually specific/contentful words (i.e. the rarer, high–tf–idf tokens) in a passage.

    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
     typelib
    -1.09
     レビュー
    -1.08
     héroe
    -1.05
    をしている
    -1.05
    these
    -1.05
    Should
    -1.05
    pherds
    -1.04
    endalikan
    -1.02
    彞
    -1.01
     Doesn
    -1.00
    POSITIVE LOGITS
    糞
    1.11
    メートル
    1.09
     setTotal
    1.02
    щал
    0.98
    蹿
    0.94
    阆
    0.93
     høre
    0.93
    ~~~~~
    0.93
     män
    0.92
    キロ
    0.90
    Activations Density 0.474%

    No Known Activations