© 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. 48337
    Prev
    Next
    INDEX
    Explanations

    without further

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

    This neuron fires on discourse-structuring adverbs that signal continuation or extension (e.g. “further,” “still,” “step” in “step further,” as in “without further ado”).

    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
     hilos
    -0.97
    Mientras
    -0.94
    ρισσ
    -0.89
    Forrás
    -0.89
     italianos
    -0.87
    También
    -0.86
    phòng
    -0.85
    Estos
    -0.83
    fungsi
    -0.83
     métal
    -0.83
    POSITIVE LOGITS
     further
    9.63
    further
    6.81
    Further
    6.81
     Further
    6.50
     FURTHER
    5.69
    进一步
    5.22
     дальше
    4.53
    FURTHER
    4.53
    さらに
    4.50
     weiter
    4.47
    Activations Density 0.337%

    No Known Activations