© 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. 22-GEMMASCOPE-RES-131K
    7. 25783
    Prev
    Next
    INDEX
    Explanations

    anise, balan, radioactive, testosterone, therapeutic

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

    chemical compounds and substances in scientific or technical descriptions.

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

    The neuron fires on densely technical or scientific noun phrases—especially formal biomedical or chemical terminology.

    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
     and
    -1.37
    existsSync
    -1.27
     вдоль
    -1.20
    頜
    -1.19
     PyLong
    -1.16
     if
    -1.12
    greenrobot
    -1.11
    姦
    -1.10
     nếu
    -1.10
     for
    -1.10
    POSITIVE LOGITS
     lampa
    1.48
     varier
    1.43
     soigneusement
    1.40
     balon
    1.32
     bici
    1.31
     palav
    1.30
     auk
    1.27
     cincin
    1.27
     perpé
    1.24
    Théâtre
    1.24
    Activations Density 0.095%

    No Known Activations