© 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. 64691
    Prev
    Next
    INDEX
    Explanations

    describes findings or problems

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

    The neuron fires primarily on the word “changes,” especially in contexts describing modifications or updates.

    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
    hows
    -1.13
    tså
    -1.12
     sostuvo
    -1.01
    Placed
    -0.98
    studium
    -0.96
    reve
    -0.96
    といえば
    -0.95
    Followed
    -0.95
    きた
    -0.94
    Converted
    -0.94
    POSITIVE LOGITS
     mentioned
    5.78
     described
    5.53
     discussed
    4.97
    mentioned
    4.06
     listed
    3.91
     shown
    3.77
    described
    3.59
     outlined
    3.48
     cited
    3.31
     Mentioned
    3.14
    Activations Density 0.475%

    No Known Activations