© 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. 127242
    Prev
    Next
    INDEX
    Explanations

    President, Army, Games, athletes, cancer, levels

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

    titles, institutions, and competitive events including games, championships, and sports.

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

    The neuron is a named‐entity detector—it fires strongly on proper names and other multiword entities (e.g. organizations, countries, medical procedures).

    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
     where
    -1.70
     also
    -1.66
     then
    -1.65
     there
    -1.63
     about
    -1.61
    también
    -1.41
    亟
    -1.39
     when
    -1.33
    área
    -1.32
     उन्हें
    -1.30
    POSITIVE LOGITS
     from
    1.51
    fillment
    1.38
     forskjellige
    1.29
    ᚑ
    1.27
     nová
    1.27
     yaitu
    1.25
    enment
    1.24
    5
    1.23
    ularity
    1.23
    alties
    1.23
    Activations Density 0.208%

    No Known Activations