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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 104251
    Prev
    Next
    INDEX
    Explanations

    What’s, reward circuitry, introduction, influence, whether

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

    The neuron activates on long hexadecimal strings (commit hashes or diff index identifiers) in version‐control diffs.

    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
     tahun
    -1.41
    )};
    -1.34
    [];
    
    -1.25
     turista
    -1.16
     ministra
    -1.16
     motocicleta
    -1.14
     asfal
    -1.11
    }));
    
    -1.11
     negara
    -1.10
     silueta
    -1.09
    POSITIVE LOGITS
    3
    1.31
    7
    1.31
    9
    1.23
    6
    1.21
    athus
    1.19
    5
    1.17
    8
    1.10
     angegebenen
    1.09
     gelernt
    1.09
    4
    1.09
    Activations Density 0.001%

    No Known Activations