© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 102611
    Prev
    Next
    INDEX
    Explanations

    starts with la

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

    The neuron is basically a detector for the two‐letter token “la.” Whenever the token “la” appears (as its own piece in the tokenizer’s output), this neuron activates strongly.

    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
    quelize
    -2.72
     ボトル
    -2.63
     своем
    -2.48
     моя
    -2.48
    CHREIB
    -2.47
     мои
    -2.45
    yyj
    -2.38
     蔬菜
    -2.38
    噐
    -2.33
    </h3>
    -2.23
    POSITIVE LOGITS
    9
    2.97
    0
    2.78
     '
    2.61
    8
    2.59
    jenige
    2.59
    </b>
    2.56
     adorable
    2.50
    1
    2.44
    3
    2.39
    a
    2.34
    Activations Density 0.020%

    No Known Activations