© 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. 63462
    Prev
    Next
    INDEX
    Explanations

    sickle cell

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

    names or terms that contain "sik" or "sick", including "sickle" in medical contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron consistently fires on subword pieces containing the character sequence “sik,” e.g. in names or words like “Sikivu,” “Sickle,” or “Sickles.”

    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
    jmniej
    -0.71
     esfor
    -0.70
     spared
    -0.69
     Nebu
    -0.68
     preliminar
    -0.68
     concurrently
    -0.67
     clearer
    -0.67
    vied
    -0.66
     grà
    -0.65
     sper
    -0.63
    POSITIVE LOGITS
    stus
    0.93
    Sick
    0.85
     Sik
    0.84
     sick
    0.84
     Sick
    0.84
    thello
    0.84
    Sic
    0.82
     Sic
    0.82
    eben
    0.80
    sika
    0.78
    Activations Density 0.027%

    No Known Activations