© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103192
    Prev
    Next
    INDEX
    Explanations

    hexadecimal characters

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

    hexadecimal characters in code or technical data structures.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron activates on long hexadecimal identifiers—i.e. strings of hex digits (0–9, a–f, A–F) such as commit hashes, GUIDs, or hex constants.

    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
     myſelf
    -2.23
    缫
    -2.16
     overwhelmingly
    -2.14
    銪
    -2.05
     casada
    -2.03
    絎
    -2.03
     thuy
    -2.02
     laatste
    -1.96
     protocolos
    -1.95
    鐗
    -1.95
    POSITIVE LOGITS
    8
    3.00
    7
    2.81
    3
    2.69
    9
    2.66
    1
    2.56
    媄
    2.34
                
    2.28
    6
    2.27
    0
    2.25
    
    2.23
    Activations Density 0.004%

    No Known Activations