© 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. 72620
    Prev
    Next
    INDEX
    Explanations

    method returns types

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

    The neuron fires on occurrences of “return” (and “returns”)—in both code statements and doc-comment return annotations.

    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
            
    -1.43
        
    -1.38
     еще
    -1.29
    🢒
    -1.27
    habited
    -1.27
     getItemCount
    -1.25
    有时
    -1.24
     ещё
    -1.24
     religiosas
    -1.23
     finner
    -1.21
    POSITIVE LOGITS
     all
    1.55
     similar
    1.34
     these
    1.32
    唷
    1.26
     size
    1.25
    ὸ
    1.23
     actionTypes
    1.21
     erweitert
    1.21
     actual
    1.20
    artige
    1.16
    Activations Density 0.004%

    No Known Activations