© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7727
    Prev
    Next
    INDEX
    Explanations

    contextual states and transitions

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

    The neuron strongly responds to terms describing covert operations—especially words about infiltration, espionage, assignments, and related mission/preparation actions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ּוֹ
    -2.00
    摂取
    -1.91
    𝒙
    -1.85
     Polícia
    -1.70
    peka
    -1.68
    pography
    -1.67
     美味
    -1.65
    💽
    -1.61
     bandoulière
    -1.60
    nước
    -1.57
    POSITIVE LOGITS
     braço
    1.94
     acidente
    1.84
     exemplo
    1.79
     relógio
    1.77
     caráter
    1.76
     brilho
    1.75
     abraço
    1.73
    ṝ
    1.72
     silêncio
    1.71
     avião
    1.70
    Activations Density 0.484%

    No Known Activations