© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 0-GEMMASCOPE-RES-65K
    11. 13406
    Prev
    Next
    INDEX
    Explanations

    the word "impermissible" and related terms indicating prohibition or restriction

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.0.hook_resid_post
    Hook Layer
    0
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     Ehren
    -0.53
    gal
    -0.52
    Origins
    -0.51
    ɨ
    -0.51
     nguồn
    -0.51
     sufrido
    -0.50
    liothèque
    -0.48
     עי
    -0.48
    vician
    -0.48
     diaspora
    -0.47
    POSITIVE LOGITS
     дописавши
    0.82
     виправивши
    0.77
     imper
    0.75
    SharedDtor
    0.74
     Dict
    0.73
    Jegyzetek
    0.72
     dict
    0.70
    itesse
    0.69
     Imper
    0.68
     dictate
    0.68
    Activations Density 0.081%

    No Known Activations