© 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. 34-GEMMASCOPE-RES-131K
    7. 81392
    Prev
    Next
    INDEX
    Explanations

    genetic study, coping, Shimerian

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

    Unicode character encodings and hexadecimal representations.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on the hexadecimal‐digit sequences used in Unicode or code‐point escape notations (e.g. “0x…”, “\u…”, “U+…”).

    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
    What
    -0.81
    主播
    -0.73
    孽
    -0.70
    lak
    -0.69
     sbag
    -0.69
    tivo
    -0.69
    flex
    -0.68
    ASER
    -0.68
     accumulating
    -0.68
    语文
    -0.68
    POSITIVE LOGITS
    Alike
    0.88
    НОСТИ
    0.85
    ండ
    0.84
     جوش
    0.83
    armament
    0.82
    hihi
    0.81
     لوله
    0.80
    toArray
    0.80
     Devin
    0.79
    ㅑ
    0.79
    Activations Density 0.009%

    No Known Activations