© 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. 22-GEMMASCOPE-RES-131K
    7. 73086
    Prev
    Next
    INDEX
    Explanations

    always seems

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

    The neuron is simply firing on very common, high‐frequency tokens (basic function words and everyday vocabulary) rather than on rare or specialized terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     dokonce
    -1.19
    くれた
    -1.08
     was
    -1.05
    beitet
    -0.99
    甚至
    -0.98
     durchaus
    -0.96
    owany
    -0.94
    CMAKE
    -0.92
    いえば
    -0.91
     همچنین
    -0.90
    POSITIVE LOGITS
     always
    1.81
     vždy
    1.56
     zawsze
    1.54
     seems
    1.53
     siempre
    1.52
    每次
    1.52
     seem
    1.49
     sempre
    1.48
     всегда
    1.45
    always
    1.45
    Activations Density 0.082%

    No Known Activations