© 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. 116694
    Prev
    Next
    INDEX
    Explanations

    the word "with" indicating connections or associations in a text

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    start user / Sithon

    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
    Negative Logits
     ſch
    -0.80
     Reſ
    -0.72
    出版年
    -0.70
     myſelf
    -0.69
     Inſ
    -0.69
     nonUne
    -0.68
     ſche
    -0.68
    })*/
    -0.67
     unſ
    -0.66
     queſta
    -0.66
    POSITIVE LOGITS
    enumi
    0.89
    enumii
    0.65
     consultato
    0.51
     
    0.45
     <<<<<<<<<<<<<<
    0.45
     the
    0.42
     vorige
    0.41
     respiración
    0.39
     fizer
    0.39
     ***!
    0.38
    Activations Density 0.000%

    No Known Activations