© 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. 32006
    Prev
    Next
    INDEX
    Explanations

    behind the scenes bts bloopers

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

    The neuron fires on words that appear in document titles or headlines (i.e. the post’s heading and sub-heading 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
     inferiores
    -0.75
    getragen
    -0.73
    игры
    -0.73
     groet
    -0.73
     Darstellung
    -0.70
     аккумуля
    -0.70
    prisingly
    -0.68
    委
    -0.68
    feira
    -0.67
     fairytale
    -0.66
    POSITIVE LOGITS
     behind
    2.67
     blo
    2.45
    Behind
    2.19
     Behind
    2.17
     deleted
    2.11
    behind
    2.06
     BTS
    1.92
    blo
    1.90
    deleted
    1.80
     Blo
    1.76
    Activations Density 0.036%

    No Known Activations