© 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. 2829
    Prev
    Next
    INDEX
    Explanations

    Hick or Heck or Hik or Hij sound

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

    The neuron activates on uppercase sequences beginning with “H” (e.g. names like Hick, Heck, Hing and acronyms like HIC), i.e. capitalized tokens starting with H.

    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
     complementa
    -0.79
    sku
    -0.75
     исследования
    -0.74
     reprodu
    -0.73
     مساء
    -0.72
     Prosperity
    -0.71
    
    -0.71
     anon
    -0.71
     oeste
    -0.71
     defer
    -0.70
    POSITIVE LOGITS
    áctenos
    0.82
    ORY
    0.80
     Hij
    0.77
     Ahi
    0.77
    clicking
    0.75
     urme
    0.75
     moeilijk
    0.75
     اتحاد
    0.74
     Had
    0.74
     Hic
    0.73
    Activations Density 0.008%

    No Known Activations