© 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. 78995
    Prev
    Next
    INDEX
    Explanations

    starts with Ras or Rash

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

    The neuron responds to occurrences of the character sequence “Ras” at the start of tokens (i.e. words beginning with “Ras”).

    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
    逓
    -0.82
    àu
    -0.78
     something
    -0.77
    tsy
    -0.74
    겐
    -0.73
     吗
    -0.73
    牡蠣
    -0.72
     greatly
    -0.70
    墾
    -0.70
    Done
    -0.69
    POSITIVE LOGITS
    bashrc
    0.85
    ाय
    0.80
     Ras
    0.79
    尸
    0.75
    heed
    0.74
    Ras
    0.73
    bora
    0.73
     RAS
    0.73
     finalist
    0.73
     مكان
    0.71
    Activations Density 0.014%

    No Known Activations