© 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. 22-GEMMASCOPE-RES-131K
    7. 15806
    Prev
    Next
    INDEX
    Explanations

    skipping build

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

    The neuron is highly activated by programming‐related tokens—code keywords, identifiers, file paths, namespaces, and other code‐snippet elements.

    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
     мост
    -1.02
     lecteur
    -0.95
    athers
    -0.94
     tôt
    -0.94
     letzten
    -0.93
    ²+
    -0.91
     laag
    -0.90
    ﭼ
    -0.90
     télécharge
    -0.90
     člov
    -0.89
    POSITIVE LOGITS
     could
    1.02
     stolet
    1.02
     other
    0.98
     periodically
    0.96
     particularly
    0.94
     for
    0.93
     extraordinary
    0.93
     or
    0.92
    When
    0.91
     frequently
    0.90
    Activations Density 0.133%

    No Known Activations