© 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. 78170
    Prev
    Next
    INDEX
    Explanations

    log followed by common suffixes

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

    The neuron strongly activates on any occurrence of the character sequence “log” (e.g. in tokens like “log,” “logmein,” “LogNEC,” “log4j,” “Logan,” 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
     האח
    -0.79
     canceling
    -0.74
    interfaces
    -0.73
     stimmt
    -0.72
    指揮
    -0.71
     Kato
    -0.70
     except
    -0.70
    ouvoir
    -0.69
    返し
    -0.69
     unre
    -0.69
    POSITIVE LOGITS
     óculos
    0.85
    ҡ
    0.84
    ほら
    0.84
    Translator
    0.82
    atle
    0.82
    rolling
    0.81
    jam
    0.81
    佷
    0.79
    是想
    0.79
     lodging
    0.79
    Activations Density 0.014%

    No Known Activations