© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 66424
    Prev
    Next
    INDEX
    Explanations

    directed coherence

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

    The neuron activates on numeric tokens or digit sequences (i.e. number literals) in code or markup.

    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
     here
    -0.81
     Jul
    -0.79
     aqui
    -0.79
     ½
    -0.78
     Aug
    -0.78
     certamente
    -0.78
     Спасибо
    -0.77
    estro
    -0.77
    ets
    -0.77
    mfenced
    -0.76
    POSITIVE LOGITS
    PSG
    0.79
    ẳng
    0.77
     forskjellige
    0.74
    ynthia
    0.74
    Ayrıca
    0.74
    Практи
    0.73
    лики
    0.73
    齦
    0.72
     különböz
    0.72
    trar
    0.72
    Activations Density 0.004%

    No Known Activations