© 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. 89880
    Prev
    Next
    INDEX
    Explanations

    ras-independent activation

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

    The neuron detects names with specific consonant patterns, particularly focusing on syllables like "Al", "ras", "sch", "ig", "pb" that appear in various proper names and scientific terms.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on occurrences of the proper name “Raschig” (as in the Raschig–Hooker process).

    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
    postId
    -0.81
    股
    -0.73
     pave
    -0.70
    DBHelper
    -0.69
     biasa
    -0.69
     sinte
    -0.68
    Pró
    -0.68
    담
    -0.67
     Newer
    -0.67
     luyện
    -0.67
    POSITIVE LOGITS
     moedas
    0.83
    tized
    0.78
    ]){
    
    0.77
    五六
    0.76
     riego
    0.75
     Nielsen
    0.74
    イケ
    0.73
    ╋
    0.72
    lion
    0.71
    nev
    0.71
    Activations Density 0.043%

    No Known Activations