© 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. 10-GEMMASCOPE-RES-131K
    4. 47760
    Prev
    Next
    INDEX
    Explanations

    purpose or recipient

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

    The neuron fires on positive evaluative adjectives (especially comparatives and superlatives) used in product descriptions to promote or praise items.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     it
    -2.13
     algunas
    -1.52
     proposent
    -1.28
     Einige
    -1.26
     menangis
    -1.24
     suivent
    -1.22
     alguns
    -1.21
    oughby
    -1.20
     skriver
    -1.19
     contextLoads
    -1.19
    POSITIVE LOGITS
     for
    2.28
    کہ
    1.72
     "'";
    1.60
    </em>
    1.52
    mesin
    1.48
     also
    1.48
    儛
    1.45
    สำหรับ
    1.42
    ствовали
    1.42
    ὸν
    1.41
    Activations Density 0.036%

    No Known Activations