© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 128760
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on the “using UnityEngine;” import line, i.e. it’s looking for UnityEngine namespace import statements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    Postprandial or there are

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    providedIn
    -1.41
    																								
    -1.38
     polisi
    -1.25
    																		
    -1.21
     }));
    -1.20
    centerTitle
    -1.20
    ЕРИ
    -1.18
     ]);
    -1.15
     any
    -1.12
     {
    
    
    -1.11
    POSITIVE LOGITS
    ↵↵
    1.21
     and
    1.18
     suiker
    1.09
     that
    1.09
    ↵
    1.06
    铿
    1.02
     diputado
    1.02
    に戻る
    0.98
    oa̍t
    0.97
     gestartet
    0.97
    Activations Density 0.001%

    No Known Activations