© 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. 103706
    Prev
    Next
    INDEX
    Explanations

    predicting probabilities

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

    The neuron is highly responsive to programming code tokens (e.g. class or def keywords, method names like fit/predict, variable names, numeric literals) and ignores plain natural‐language text.

    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
     Predictions
    -0.82
    toDate
    -0.82
     predictions
    -0.81
    Prediction
    -0.79
    corpo
    -0.78
    macar
    -0.75
    predictions
    -0.75
     Reverend
    -0.73
    drawString
    -0.73
    中共
    -0.72
    POSITIVE LOGITS
    proba
    1.10
     on
    1.02
     EH
    0.82
     probabilistic
    0.82
    inkler
    0.80
     probabilities
    0.75
     appartient
    0.74
     вероят
    0.74
     onstage
    0.73
    Pollution
    0.73
    Activations Density 0.011%

    No Known Activations