© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 47619
    Prev
    Next
    INDEX
    Explanations

    contrastive conjunctions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     if
    -1.41
     must
    -1.17
     certainly
    -1.16
     especially
    -1.13
     justru
    -1.09
     even
    -1.08
     simply
    -1.07
     obviously
    -1.06
     malah
    -1.05
     probably
    -1.03
    POSITIVE LOGITS
    sometimes
    1.07
    嘴
    1.01
    知道
    0.96
     circunstan
    0.96
     locul
    0.95
    titre
    0.94
     pleins
    0.94
     partea
    0.94
     virker
    0.94
    quently
    0.93
    Activations Density 0.009%

    No Known Activations