© 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. 34-GEMMASCOPE-RES-131K
    4. 34766
    Prev
    Next
    INDEX
    Explanations

    Biol or information

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

    open source license and copyright headers in code files.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron responds to tokens that contain underscores (e.g. resource or file‐path names like “simple_spinner_item,” “group_lifecycle_policy,” 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
     Bezirks
    -0.73
     alive
    -0.71
     lige
    -0.69
    vode
    -0.67
    invasive
    -0.66
     immersive
    -0.66
    Atom
    -0.66
     explaining
    -0.66
    thyst
    -0.64
    cemia
    -0.64
    POSITIVE LOGITS
     Buchanan
    0.69
     wildest
    0.69
    heal
    0.66
    buen
    0.64
    qm
    0.63
     Godfrey
    0.63
     "---
    0.61
    czyna
    0.60
     "\"
    0.60
     Sho
    0.60
    Activations Density 0.083%

    No Known Activations