© 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. 60857
    Prev
    Next
    INDEX
    Explanations

    reproduction of media

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

    The neuron detects the dense technical jargon of patent descriptions—especially terms around data storage, recording/producing apparatus, and related invention‐section language.

    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
     Hemisphere
    -0.81
    essi
    -0.74
    siyon
    -0.69
    icto
    -0.68
    hljs
    -0.66
    boli
    -0.66
     frecuentes
    -0.66
     Pfer
    -0.66
     handic
    -0.65
     kỷ
    -0.65
    POSITIVE LOGITS
     playback
    4.84
     play
    3.61
     Playback
    3.53
    playback
    3.52
     playing
    3.34
    Playback
    3.34
     played
    3.25
    播放
    3.25
    再生
    3.00
    play
    2.84
    Activations Density 0.172%

    No Known Activations