© 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-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 27068
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "ein" and its variations in German text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    a, an, ein, un

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     honneur
    -0.66
     trône
    -0.54
     identité
    -0.54
     vôtre
    -0.49
     lendemain
    -0.49
     delà
    -0.47
     tarko
    -0.47
    paravant
    -0.45
     vêtement
    -0.45
     espoir
    -0.44
    POSITIVE LOGITS
     a
    1.00
     an
    0.95
    niſſe
    0.90
     ஒரு
    0.81
     another
    0.81
    ſicht
    0.79
     একটি
    0.79
    ſſung
    0.75
    ſehen
    0.75
     small
    0.73
    Activations Density 0.000%

    No Known Activations