© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 33013
    Prev
    Next
    INDEX
    Explanations

    nabu, nab, gab

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

    The neuron consistently activates on any token containing the character sequence “Nab,” i.e. names or abbreviations beginning with “Nab.”

    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
     satisfa
    -0.84
    хни
    -0.77
     belir
    -0.73
     Emanuel
    -0.72
    仔细
    -0.72
     طريقة
    -0.71
    CONTINUE
    -0.70
     ases
    -0.70
     calcular
    -0.70
     António
    -0.69
    POSITIVE LOGITS
     NAB
    1.07
    NAB
    0.93
    Nab
    0.87
    łek
    0.85
     PublicKey
    0.76
     nab
    0.75
    nab
    0.75
     eczema
    0.74
    ַת
    0.73
    ěl
    0.73
    Activations Density 0.024%

    No Known Activations