© 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. 34-GEMMASCOPE-RES-131K
    7. 99673
    Prev
    Next
    INDEX
    Explanations

    creating threads and topics

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

    The neuron fires on standalone title‐style headings or section headers—especially those containing uppercase words and numbers (e.g. “Official 2016 Playoffs Bandwagon Guide”).

    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
    わが
    -0.76
     voegen
    -0.73
    했다
    -0.71
     And
    -0.70
    rijfs
    -0.68
    arrangement
    -0.68
    UnityEngine
    -0.67
     No
    -0.67
    시아
    -0.66
    ᴸ
    -0.66
    POSITIVE LOGITS
     thread
    6.44
     threads
    5.31
    thread
    4.91
     Thread
    4.44
    Thread
    4.25
     THREAD
    4.03
    threads
    3.73
    线程
    3.56
     Threads
    3.53
    THREAD
    3.52
    Activations Density 0.109%

    No Known Activations