© 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. 16218
    Prev
    Next
    INDEX
    Explanations

    medical or legal continuations

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

    text in legal or license statements, particularly around version numbers and comma-or phrases.

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

    The neuron responds to standalone numeric tokens, particularly numbers used in citations, measurements, or other quantitative references.

    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
     roj
    -0.96
     pasto
    -0.93
     realizzata
    -0.90
    RLock
    -0.89
     gaun
    -0.87
    Поэтому
    -0.87
     maso
    -0.84
     kuri
    -0.83
     kafe
    -0.83
     jaj
    -0.82
    POSITIVE LOGITS
    及其
    1.05
     dingen
    1.02
     ainda
    0.90
     plus
    0.88
     плюс
    0.88
     etiam
    0.88
     případ
    0.87
     vēl
    0.85
     tenía
    0.85
     depois
    0.85
    Activations Density 0.081%

    No Known Activations