© 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. 22-GEMMASCOPE-RES-131K
    7. 26907
    Prev
    Next
    INDEX
    Explanations

    let $

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

    This neuron fires on the little section‐header or step‐intro markers (e.g. “HINT:”, “A:”, “Let …”) that introduce a solution or hint in a math write-up.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     групи
    -0.90
    わかった
    -0.88
     '_
    -0.85
    ilogue
    -0.84
     however
    -0.84
     couple
    -0.83
     themſelves
    -0.83
    /%
    -0.81
    ּוֹ
    -0.81
     recommend
    -0.80
    POSITIVE LOGITS
    뤄
    1.03
     autre
    1.01
    Comté
    0.99
    ικοί
    0.97
    withIdentifier
    0.96
     jums
    0.93
    ឵
    0.93
    ⍽
    0.93
    𖤍
    0.93
    ⭓
    0.93
    Activations Density 0.009%

    No Known Activations