© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 62602
    Prev
    Next
    INDEX
    Explanations

    reaching or arriving

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

    The neuron detects tokens characteristic of step‐by‐step instructions or directions (e.g. “you,” “will,” “follow,” “keep,” “after,” numeric cues) in procedural or navigational text.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    suchen
    -0.87
    formas
    -0.85
    went
    -0.83
    Went
    -0.83
     beper
    -0.82
    agisse
    -0.82
     pergi
    -0.81
    𓆏
    -0.81
    jalan
    -0.81
    шения
    -0.80
    POSITIVE LOGITS
     reach
    3.13
     reached
    3.05
     reaches
    2.69
     reaching
    2.67
     arrived
    2.52
     arrive
    2.48
     Reach
    2.23
    reach
    2.20
    Reach
    2.16
    Reaching
    2.14
    Activations Density 0.035%

    No Known Activations