© 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. 111865
    Prev
    Next
    INDEX
    Explanations

    as for

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

    The neuron fires on occurrences of the discourse‐marking bigram “as for.”

    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
    </td>
    -1.86
     queste
    -1.84
     between
    -1.80
    </h5>
    -1.74
    でも
    -1.70
     dieser
    -1.70
     freaking
    -1.69
     With
    -1.66
    这些
    -1.66
    "]
    -1.55
    POSITIVE LOGITS
    卨
    1.98
    咔嚓
    1.88
     them
    1.85
    秾
    1.70
     didik
    1.69
    ⛦
    1.68
     raiſ
    1.65
    噼
    1.63
     betek
    1.58
    setTypeface
    1.54
    Activations Density 0.002%

    No Known Activations