© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 77452
    Prev
    Next
    INDEX
    Explanations

    victim or target

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

    The neuron strongly activates on very common function words, punctuation, and other high-frequency 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
    itoriale
    -0.82
    Topping
    -0.81
    operable
    -0.81
     postępo
    -0.81
     angelegt
    -0.81
    Probable
    -0.81
     parha
    -0.79
    mortem
    -0.76
     τῶν
    -0.75
     mahd
    -0.75
    POSITIVE LOGITS
    被
    1.70
     target
    1.48
    Target
    1.19
     被
    1.17
    TARGET
    1.14
     victim
    1.12
    target
    1.11
    是被
    1.04
     Target
    1.02
    setTarget
    1.01
    Activations Density 0.197%

    No Known Activations