© 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. 103701
    Prev
    Next
    INDEX
    Explanations

    habitat destruction

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

    text related to government, institutional, or organizational information.

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

    The neuron strongly activates on numeric tokens and number‐heavy sequences (e.g. dates, measurements, counts).

    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
    Captcha
    -0.85
    ステル
    -0.79
    Syria
    -0.73
    一楼
    -0.72
    روان
    -0.71
    terape
    -0.70
     pabrik
    -0.69
    initialized
    -0.67
    вості
    -0.66
     glad
    -0.66
    POSITIVE LOGITS
     endangered
    0.97
     EP
    0.92
     extinction
    0.91
     federal
    0.83
     TRIG
    0.81
    ={[
    0.79
     endanger
    0.79
    ppone
    0.78
     offsets
    0.77
    oltà
    0.75
    Activations Density 0.004%

    No Known Activations