© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 118315
    Prev
    Next
    INDEX
    Explanations

    words or short phrases

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

    This neuron detects phonetic transcription markers—that is, the slash‐delimited phoneme labels (e.g. “/da/,” “/ta/”) used to indicate phonetic categories.

    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
    Să
    -0.98
     cérebro
    -0.91
    Primeiro
    -0.88
    énieurs
    -0.85
     genç
    -0.85
     ensaio
    -0.83
    ∝
    -0.82
    ナイキ
    -0.82
     reager
    -0.82
    Boiler
    -0.82
    POSITIVE LOGITS
     father
    1.00
     boat
    0.94
     water
    0.94
     night
    0.93
     bake
    0.90
     house
    0.89
     river
    0.87
     fish
    0.86
     ship
    0.86
     clothes
    0.86
    Activations Density 0.030%

    No Known Activations