© 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. 102645
    Prev
    Next
    INDEX
    Explanations

    press release or news headlines

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

    The neuron strongly activates on tokens that are purely numeric (or numeric sequences) and adjacent punctuation (like &, –, colons), i.e. it detects number‐like tokens.

    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
     half
    -0.70
    fillable
    -0.69
    nall
    -0.69
    Indiana
    -0.68
     населения
    -0.68
     στοι
    -0.68
     reproach
    -0.66
     Bye
    -0.66
     OnTrigger
    -0.66
     sukienka
    -0.66
    POSITIVE LOGITS
    FRAME
    0.96
     riders
    0.93
    ,''
    0.87
     rider
    0.85
     ASTM
    0.82
    framed
    0.81
     FRAME
    0.81
     frame
    0.80
    frame
    0.78
     🙏
    0.77
    Activations Density 0.005%

    No Known Activations