© 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. 75551
    Prev
    Next
    INDEX
    Explanations

    snapped losing streaks

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

    This neuron detects verbs (and nearby articles/numerals) that describe ending a run or streak—e.g. “snapped a three-game losing streak,” “broke a nation’s heart,” “break Mayor Ford’s stranglehold.”

    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
     a
    -1.51
    じゃなくて
    -1.30
     marrom
    -1.28
    ⢅
    -1.22
     égard
    -1.22
     amarela
    -1.20
    UCHAR
    -1.19
    冂
    -1.19
    ジャケット
    -1.16
    はこんな感じ
    -1.16
    POSITIVE LOGITS
     have
    1.39
    PLEASE
    1.27
    ❌
    1.27
     一般
    1.27
    bemos
    1.23
    การ
    1.23
     Comando
    1.23
           
    1.23
    青森
    1.22
    it
    1.20
    Activations Density 0.067%

    No Known Activations