© 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. 20953
    Prev
    Next
    INDEX
    Explanations

    DSM or bgcolor

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

    The neuron lights up on technical or domain‐specific identifiers—tokens like class names (e.g. “JFrame”), configuration keywords (e.g. “bgcolor”), or uppercase acronyms.

    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
    ⽌
    -3.23
    瑨
    -3.08
    昻
    -3.06
    ⤥
    -3.05
    ୩
    -2.98
    ༀ
    -2.81
    
    -2.67
    神
    -2.66
    玧
    -2.64
    蠓
    -2.61
    POSITIVE LOGITS
    \
    3.55
    ve
    3.22
    tu
    3.08
    st
    3.06
     插画
    2.86
    O
    2.84
    ac
    2.84
    er
    2.83
    "
    2.75
    )
    2.73
    Activations Density 0.007%

    No Known Activations