© 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. 10-GEMMASCOPE-RES-131K
    7. 112420
    Prev
    Next
    INDEX
    Explanations

    codeigniter, ignorable, ignoring

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

    the letter sequence "ign" appearing in technical or proper nouns, particularly in words like "Codeigniter", "Verisign", "malignancy", and "IGN".

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    al
    -2.81
     рассказать
    -2.75
    This
    -2.72
    While
    -2.66
    These
    -2.64
    However
    -2.64
    With
    -2.61
    ной
    -2.59
    er
    -2.47
     emphasizes
    -2.45
    POSITIVE LOGITS
    ſſen
    3.52
    ৢ
    2.94
     they
    2.88
    栳
    2.78
     ausein
    2.78
    
    2.73
     ſch
    2.72
    uccess
    2.69
    rugs
    2.69
    鹛
    2.67
    Activations Density 0.006%

    No Known Activations