© 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. 45040
    Prev
    Next
    INDEX
    Explanations

    null and separators

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

    The neuron fires on occurrences of the literal “null,” i.e. it detects null-checks or null references in code.

    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
    かっこ
    -1.61
    　　-
    -1.59
     decorativa
    -1.55
    勿論
    -1.54
     pomá
    -1.47
     駅前
    -1.46
    さあ
    -1.44
     汽車
    -1.41
    　　「
    -1.37
    ΩΝ
    -1.36
    POSITIVE LOGITS
    “
    1.64
     I
    1.52
    
    1.52
    }
    1.45
    \
    1.39
     they
    1.38
    —.
    1.38
     Luego
    1.34
     uomini
    1.34
     affichés
    1.31
    Activations Density 0.002%

    No Known Activations