© 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. 8665
    Prev
    Next
    INDEX
    Explanations

    heap data structure

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

    The neuron is specialized to detect the word “heap” (and its close morphological variants) in technical or code-related contexts.

    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
    錞
    -2.67
    ፨
    -2.67
    ":"+
    -2.58
    椟
    -2.55
    𖤣
    -2.53
    ︕
    -2.50
    鉭
    -2.48
    釹
    -2.47
    ⫘
    -2.47
    .”
    -2.45
    POSITIVE LOGITS
    '
    3.33
     
    2.86
     '
    2.81
     reconoc
    2.48
    <bos>
    2.47
     recuerdo
    2.38
    ...
    2.33
    {
    2.33
    </strong>
    2.28
    ]
    2.28
    Activations Density 0.001%

    No Known Activations