© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 75171
    Prev
    Next
    INDEX
    Explanations

    code identifiers

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

    The neuron consistently activates on code identifiers or type names that begin with uppercase letters (e.g. VM, MemPool, THREE, Context, State).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     another
    -0.89
    واف
    -0.79
     that
    -0.79
     think
    -0.78
    
    -0.78
    기를
    -0.77
     more
    -0.74
    報道
    -0.73
     each
    -0.73
     on
    -0.73
    POSITIVE LOGITS
     époux
    0.96
     paramètres
    0.85
     inton
    0.81
    찹
    0.81
     excru
    0.81
     BOB
    0.80
     toegang
    0.80
    всем
    0.79
     jonge
    0.79
     conozco
    0.79
    Activations Density 0.036%

    No Known Activations