© 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. 34-GEMMASCOPE-RES-131K
    7. 58361
    Prev
    Next
    INDEX
    Explanations

    construction and material measurements

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

    The neuron is specialized to pick up numeric dimension and measurement tokens (e.g. board sizes like “2 x 4,” lengths in inches, 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
     tubes
    -0.89
     tubular
    -0.86
     panes
    -0.83
     Dian
    -0.79
    Pró
    -0.77
     pipes
    -0.73
    pipes
    -0.71
    Hc
    -0.71
     dilated
    -0.70
    iency
    -0.69
    POSITIVE LOGITS
     lumber
    0.89
     scraps
    0.87
     материал
    0.82
     pine
    0.79
     embalaje
    0.73
    Annex
    0.72
    Rough
    0.72
     пишет
    0.71
     McCoy
    0.71
     Archivo
    0.71
    Activations Density 0.024%

    No Known Activations