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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 65311
    Prev
    Next
    INDEX
    Explanations

    python for loops and ranges

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

    The neuron detects integer numeric literals (digits) in code, especially when they appear as function-call or indexing arguments.

    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
     as
    -1.52
     by
    -1.13
     some
    -1.11
     blatantly
    -1.10
     having
    -1.10
     these
    -1.07
     any
    -1.04
     this
    -1.04
     so
    -1.03
     many
    -1.02
    POSITIVE LOGITS
    ):
    
    1.06
     représenter
    1.05
     levou
    1.02
     différentes
    1.00
     иссле
    0.99
     différents
    0.99
    ,
    0.98
     then
    0.97
    len
    0.95
    Becoming
    0.94
    Activations Density 0.004%

    No Known Activations