© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10718
    Prev
    Next
    INDEX
    Explanations

    presents a

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find occurrences of the word “present” (including its morphological variants like presents, presented, presenting).

    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
     loto
    -1.88
    2
    -1.86
     kef
    -1.82
     kei
    -1.81
     doos
    -1.80
    鹇
    -1.80
     tremb
    -1.79
     tille
    -1.77
     lich
    -1.77
     frau
    -1.77
    POSITIVE LOGITS
     If
    2.34
     querido
    2.23
     också
    2.23
     joyas
    2.20
    </h4>
    2.20
     The
    2.19
     tomado
    2.19
     cuarent
    2.11
     sufrido
    2.06
     если
    2.03
    Activations Density 0.028%

    No Known Activations