© 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. 111066
    Prev
    Next
    INDEX
    Explanations

    begin followed by commands

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

    This neuron fires on occurrences of the token “Begin” (or its lowercase variant “begin_”), i.e. markers that start code operations or function calls.

    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
     روزی
    -1.04
    Mim
    -0.96
     vues
    -0.85
     renowned
    -0.83
     распо
    -0.82
     Mim
    -0.82
     becoming
    -0.81
    disp
    -0.81
     voisin
    -0.80
    が見える
    -0.80
    POSITIVE LOGITS
    nig
    1.15
    ners
    1.05
    inning
    1.02
    ner
    1.00
    jahr
    0.96
    ofday
    0.96
    istiques
    0.95
    licitations
    0.95
    neling
    0.94
     obyvateľov
    0.94
    Activations Density 0.016%

    No Known Activations