© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 23603
    Prev
    Next
    INDEX
    Explanations

    I as a subject

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

    This neuron tracks token position in the segment, activating most strongly on words that occur near the end.

    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
    Judging
    -1.32
    Upon
    -1.20
    Alright
    -1.20
     كافة
    -1.18
     whatnot
    -1.16
    hichever
    -1.10
    İŞ
    -1.03
     Liban
    -1.03
    Needless
    -1.02
     fraî
    -1.02
    POSITIVE LOGITS
     principally
    1.08
     maybe
    1.00
     incomes
    0.96
     aren
    0.94
     excepting
    0.90
     respect
    0.89
     don
    0.88
    Statistic
    0.88
     beside
    0.85
     supposition
    0.85
    Activations Density 0.376%

    No Known Activations