© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72555
    Prev
    Next
    INDEX
    Explanations

    release from confinement

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

    The neuron flags words and phrases describing people being released or freed from detention or prison.

    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
     тонн
    -1.15
    uncurkan
    -1.09
    繽
    -1.04
    ranton
    -1.03
     cualquier
    -1.02
     panoramique
    -1.00
    階の
    -0.98
     цены
    -0.97
    <bos>
    -0.96
    ƪ
    -0.96
    POSITIVE LOGITS
     after
    1.91
     from
    1.88
     out
    1.59
     release
    1.43
    ToReturn
    1.34
     released
    1.31
    释放
    1.30
     discharge
    1.24
     without
    1.23
     three
    1.19
    Activations Density 0.024%

    No Known Activations