© 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. 34-GEMMASCOPE-RES-131K
    7. 93447
    Prev
    Next
    INDEX
    Explanations

    sequence, seq, subsequence

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

    This neuron fires on the individual name‐and‐dot segments of fully qualified identifiers (e.g. Java package or import statements).

    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
     gone
    -0.91
    まあ
    -0.77
     stink
    -0.74
     endeavours
    -0.74
     Nepali
    -0.73
    ünk
    -0.72
    先天
    -0.72
    これは
    -0.72
    cetamol
    -0.72
     unveiling
    -0.72
    POSITIVE LOGITS
    Seq
    0.98
     seq
    0.91
     Seq
    0.88
     sequence
    0.87
    seq
    0.85
     Sequence
    0.83
    sequence
    0.81
     факторов
    0.81
     シー
    0.79
     subsequence
    0.78
    Activations Density 0.026%

    No Known Activations