© 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. 10-GEMMASCOPE-RES-131K
    7. 14376
    Prev
    Next
    INDEX
    Explanations

    go commands

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

    This neuron fires on C-style include‐guard macro names (all‐caps identifiers ending in “_H,” e.g. UD_DECODE_H, _LIB_ROOT_MEMORY_PRIVATE_H, _IPXE_RETRY_H).

    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
    ujuk
    -1.49
    adjective
    -1.42
    YMS
    -1.35
    Слайд
    -1.35
     Obten
    -1.34
     каба
    -1.33
    ECU
    -1.32
     paradox
    -1.32
     могло
    -1.31
    =-=-
    -1.30
    POSITIVE LOGITS
     everybody
    1.63
     everyone
    1.56
     everything
    1.38
     about
    1.35
     –
    1.27
     anyone
    1.23
     okay
    1.23
     zrobi
    1.23
     („
    1.23
     dlaczego
    1.23
    Activations Density 0.125%

    No Known Activations