© 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. 103266
    Prev
    Next
    INDEX
    Explanations

    places and stages for activity

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

    The neuron strongly activates on the indefinite article “a” and the content words immediately following it (i.e. the start of “a …” noun phrases).

    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
     پایه
    -0.90
    miguel
    -0.83
    jMenuItem
    -0.82
     nhóm
    -0.82
    motorola
    -0.79
    onacci
    -0.77
    录制
    -0.75
    weer
    -0.75
     paulista
    -0.74
    CONTACTS
    -0.74
    POSITIVE LOGITS
     laboratory
    1.78
     playground
    1.66
     canvas
    1.41
     classroom
    1.25
     theatre
    1.23
    playground
    1.23
     arena
    1.22
     giant
    1.19
     laboratories
    1.16
     battlefield
    1.16
    Activations Density 0.053%

    No Known Activations