© 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. 34-GEMMASCOPE-RES-131K
    7. 3169
    Prev
    Next
    INDEX
    Explanations

    recovery

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

    The neuron fires on long runs of whitespace (especially leading spaces used for indentation or formatting).

    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
    thinkable
    -0.77
     manifestazione
    -0.72
    Evans
    -0.69
     Caster
    -0.68
    mejores
    -0.68
    ובר
    -0.66
    后退
    -0.66
     سو
    -0.66
     SCIENTIFIC
    -0.65
    CCI
    -0.64
    POSITIVE LOGITS
     recovery
    0.99
    recovery
    0.98
     Recover
    0.96
     recoverable
    0.95
    Recover
    0.94
    recover
    0.94
     Recovery
    0.91
    Recovery
    0.89
     recover
    0.85
     recovered
    0.85
    Activations Density 0.023%

    No Known Activations