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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43529
    Prev
    Next
    INDEX
    Explanations

    recapit, lute, ecclesiastical

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

    The neuron responds to low‐frequency or otherwise “unusual” token fragments—i.e. rare words, uncommon subword stems, or out-of-vocabulary character sequences.

    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
     techniczne
    -2.63
    ↵↵
    -2.63
     возможность
    -2.58
    ⡪
    -2.52
    .
    -2.42
     andere
    -2.39
     cambios
    -2.33
    鶿
    -2.33
    shall
    -2.30
     megfelelő
    -2.28
    POSITIVE LOGITS
    鹮
    2.94
    i
    2.66
     Now
    2.59
     

    2.45
     Could
    2.33
     bustling
    2.33
     ominous
    2.30
     That
    2.28
     Теперь
    2.25
     Когда
    2.23
    Activations Density 0.021%

    No Known Activations