© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 115521
    Prev
    Next
    INDEX
    Explanations

    Network changes, community event, evaluation of

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

    The neuron strongly activates on numeric literals (decimal or hex) in code, effectively detecting sequences of digits (and hex digits) representing constants.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     just
    -1.12
     JUST
    -1.08
     any
    -1.00
     Just
    -0.95
     in
    -0.94
     by
    -0.90
    سع
    -0.88
    Just
    -0.86
     Before
    -0.85
    just
    -0.85
    POSITIVE LOGITS
    };
    1.13
    ];
    1.09
     };
    1.07
     captivating
    1.04
     delightfully
    1.00
     mediocre
    0.98
     shimmering
    0.98
    },{
    0.98
    0
    0.97
    },
    0.96
    Activations Density 0.002%

    No Known Activations