© 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. 34-GEMMASCOPE-RES-131K
    7. 99376
    Prev
    Next
    INDEX
    Explanations

    afrikáans, afisha, ASF licenses

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

    The neuron strongly fires on short, lowercase letter‐only tokens that look like technical codes or abbreviations (e.g. the “avh”/“dx” pieces of an “.avhdx” filename and the repeated “af” language‐code tokens).

    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
     partnering
    -0.83
    があり
    -0.75
    Tg
    -0.74
     statuses
    -0.73
     partner
    -0.73
     Berliner
    -0.72
    insch
    -0.71
     utafitiHapana
    -0.71
    _{-\
    -0.69
    tunnel
    -0.68
    POSITIVE LOGITS
     canta
    0.91
    Zá
    0.89
     terminated
    0.85
    さぁ
    0.82
    Limited
    0.80
    forhold
    0.80
    ()<<
    0.79
     Dě
    0.78
    Departamento
    0.76
     quidem
    0.76
    Activations Density 0.004%

    No Known Activations