© 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. 34-GEMMASCOPE-RES-131K
    4. 116793
    Prev
    Next
    INDEX
    Explanations

    stoked, epic, lit, rad, awesome

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

    The neuron is acting like a “common‐word” detector, giving its strongest activations to high‐frequency (very common) tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ค่ะ
    -1.06
     Atlántico
    -1.02
     pastillas
    -0.98
    UNUSED
    -0.93
     valmist
    -0.92
     منبع
    -0.92
    Πηγές
    -0.91
    cestry
    -0.90
     गर्न
    -0.90
    祂
    -0.89
    POSITIVE LOGITS
     stoked
    1.42
     epic
    1.17
     shit
    1.09
     legit
    1.09
     rad
    1.09
     dudes
    1.06
     instagram
    1.00
    gnar
    1.00
     awesome
    0.98
     shitty
    0.98
    Activations Density 0.094%

    No Known Activations