© 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. 43993
    Prev
    Next
    INDEX
    Explanations

    getting next element

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

    The neuron fires on occurrences of the identifier “next,” i.e. when code refers to a next element, pointer, iterator, or similar.

    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
    </strong>
    -2.73
     GETTING
    -1.83
     effectuer
    -1.63
     communiquer
    -1.63
    かっこいい
    -1.62
     sime
    -1.62
    把她
    -1.57
    ↵
    -1.56
    かっこ
    -1.55
     magnific
    -1.54
    POSITIVE LOGITS
     Some
    1.84
     any
    1.75
     Each
    1.71
     staunch
    1.65
     kommentar
    1.63
     relentlessly
    1.63
     EVERY
    1.61
    dzić
    1.60
     ơi
    1.60
    </h5>
    1.59
    Activations Density 0.009%

    No Known Activations