© 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. 34-GEMMASCOPE-RES-131K
    4. 119343
    Prev
    Next
    INDEX
    Explanations

    Generated by

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

    The neuron fires on isolated capitalized names or labels (proper‐noun tokens like “Google,” “Unknown,” “Bluetooth,” “Author,” or “Generated”).

    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
     ciudades
    -1.01
     ralla
    -1.00
    ęcia
    -0.96
    鯵
    -0.94
     Armee
    -0.94
     its
    -0.93
    aterin
    -0.90
    ricas
    -0.90
    遇到的
    -0.89
     Zerst
    -0.88
    POSITIVE LOGITS
    </i>
    1.21
    +:+
    1.06
    </
    1.05
    </b>
    1.00
    ELSE
    0.91
     while
    0.90
     after
    0.87
    <td>
    0.86
    找
    0.85
    </td>
    0.85
    Activations Density 0.001%

    No Known Activations