© 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. 67736
    Prev
    Next
    INDEX
    Explanations

    blessed with new things

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

    The neuron activates on words that express being blessed, honored, endowed, or otherwise receiving favor.

    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
     you
    -1.78
     everybody
    -1.76
    ]=>
    -1.63
     soorten
    -1.59
     として
    -1.58
     bekym
    -1.56
    玧
    -1.55
    ...”
    -1.52
     Viitattu
    -1.51
     gesamten
    -1.51
    POSITIVE LOGITS
     如果
    1.78
    新的
    1.68
    "
    1.61
    Until
    1.59
     这
    1.55
     trovano
    1.55
    These
    1.55
    From
    1.55
     águas
    1.55
    Since
    1.53
    Activations Density 0.009%

    No Known Activations