© 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. 38428
    Prev
    Next
    INDEX
    Explanations

    ending, terminating, eliminating

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

    This neuron activates on word stems related to ending or removal (e.g. “termin‐” and “elimin‐” in “terminable,” “terminate,” “eliminate,” etc.).

    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
    דר
    -0.81
    isyon
    -0.80
    üstü
    -0.77
    ದ
    -0.75
    unsupported
    -0.75
    }),
    -0.75
     }),
    -0.75
    思
    -0.75
    וצ
    -0.72
     Bailly
    -0.72
    POSITIVE LOGITS
    ators
    1.03
    Termination
    1.01
    Terminate
    0.99
    Elimin
    0.95
    termin
    0.94
     Termination
    0.94
     terminated
    0.94
     Elimination
    0.93
     Termin
    0.92
     terminate
    0.90
    Activations Density 0.042%

    No Known Activations