© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67472
    Prev
    Next
    INDEX
    Explanations

    Copyright notice

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

    The neuron strongly activates on the token “rem” when it appears as a standalone comment marker (e.g. in lines beginning “rem”) in source‐file headers.

    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
     besonders
    -1.04
     at
    -0.94
     or
    -0.92
    場所で
    -0.91
     dekat
    -0.84
    感じで
    -0.84
     murah
    -0.84
     unserem
    -0.83
    くて
    -0.82
    為に
    -0.82
    POSITIVE LOGITS
    がありますが
    1.02
     gestión
    0.97
     vinci
    0.97
    bibinfo
    0.96
    もありますが
    0.96
    pila
    0.92
    記憶
    0.92
    ratulations
    0.92
    ELTS
    0.92
    loren
    0.90
    Activations Density 0.093%

    No Known Activations