© 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. 116972
    Prev
    Next
    INDEX
    Explanations

    Canadian biochemist

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

    biographical information in parentheses, particularly birth and death dates or time periods a person served in a position.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The main thing this neuron does is detect numeric tokens (especially years and other standalone numbers).

    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
    嘻嘻
    -0.76
    actividad
    -0.75
    んでいます
    -0.73
    电商
    -0.73
     ajuda
    -0.71
     从
    -0.70
     jiné
    -0.70
    糍
    -0.69
    ulu
    -0.69
    dispatchEvent
    -0.69
    POSITIVE LOGITS
     predeceased
    1.32
    prefab
    0.93
     special
    0.87
     Mom
    0.84
    ৃ
    0.82
    missed
    0.80
     missed
    0.79
     will
    0.79
    外套
    0.79
     Scoring
    0.78
    Activations Density 0.006%

    No Known Activations