© 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. 34-GEMMASCOPE-RES-131K
    7. 48460
    Prev
    Next
    INDEX
    Explanations

    numbers and words starting with Q or B or P

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

    The neuron strongly activates on technical telecom‐spec vocabulary—i.e. domain-specific terms and acronyms (like “5G,” “radio,” “network,” “MHz,” “deployment scenarios,” etc.) typical of wireless communication standards documents.

    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
     LTE
    -0.81
     GSM
    -0.81
     chun
    -0.74
    dH
    -0.73
     переда
    -0.72
     cantos
    -0.72
     Ea
    -0.71
     *_
    -0.70
     молоко
    -0.70
     Avo
    -0.69
    POSITIVE LOGITS
    隈
    0.84
     méda
    0.80
     mm
    0.80
    relativeTime
    0.79
    oredCriteria
    0.78
     slicing
    0.75
     vegano
    0.75
    毫米
    0.75
     humanidad
    0.74
    жды
    0.73
    Activations Density 0.034%

    No Known Activations