© 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. 110876
    Prev
    Next
    INDEX
    Explanations

    running simulations and schematics

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

    The neuron responds to high‐index (rare) tokens—especially numerals and obscure technical identifiers.

    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
    šku
    -0.88
     blur
    -0.78
     MPC
    -0.77
     Fib
    -0.77
    spender
    -0.76
     IMA
    -0.74
    writeFieldEnd
    -0.74
    fib
    -0.73
    樵
    -0.71
     Dub
    -0.71
    POSITIVE LOGITS
    spice
    1.12
    Spice
    1.01
    simulation
    0.95
     Spice
    0.93
     simulators
    0.93
     simulator
    0.90
    Simulation
    0.89
    Sweep
    0.88
     Spectre
    0.88
     spice
    0.88
    Activations Density 0.014%

    No Known Activations