© 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. 10-GEMMASCOPE-RES-131K
    4. 124047
    Prev
    Next
    INDEX
    Explanations

    First pronoun then verb

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

    The neuron activates on the opening “First … ” in step-by-step instructions, especially the “First you need to…” (and similar first‐step) phrasing.

    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
    既に
    -1.72
    ?
    -1.69
    すでに
    -1.65
     procedentes
    -1.41
     comienzos
    -1.40
    雱
    -1.39
      
    -1.38
    –
    -1.37
    {
    -1.34
    皤
    -1.33
    POSITIVE LOGITS
     andato
    1.83
    卨
    1.75
     Which
    1.70
    }
    1.65
    şiv
    1.48
     Portugu
    1.47
     bum
    1.45
     Psico
    1.42
     hän
    1.42
    न्ह
    1.42
    Activations Density 0.017%

    No Known Activations