© 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. 22-GEMMASCOPE-RES-131K
    4. 69224
    Prev
    Next
    INDEX
    Explanations

    <?xml and code

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

    The main thing this neuron does is detect positive evaluative words expressing praise or admiration.

    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
     ilgi
    -1.12
    ର
    -1.08
    其余
    -1.08
    Nachteile
    -1.06
     mengakibatkan
    -1.05
     klachten
    -1.05
    aret
    -1.04
    seamnă
    -1.03
    修为
    -1.02
     persönliche
    -1.02
    POSITIVE LOGITS
     of
    1.21
     such
    1.15
     vandens
    1.06
     especially
    1.05
    kých
    0.97
     établissements
    0.94
    :],
    0.94
     studies
    0.94
    Blog
    0.93
    mybatisplus
    0.91
    Activations Density 0.011%

    No Known Activations