© 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
    7. Residual Stream - 16k
    9. 40-GEMMASCOPE-RES-16K
    11. 8832
    Prev
    Next
    INDEX
    Explanations

    strong emphasis on the word "very."

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 40-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_40/width_16k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.40.hook_resid_post
    Hook Layer
    40
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    IntoConstraints
    -0.49
    better
    -0.46
     الرياضيه
    -0.46
     snippetHide
    -0.45
     relativement
    -0.43
     absolument
    -0.42
     كومونز
    -0.42
     الحره
    -0.41
    Easier
    -0.41
     SwitchCompat
    -0.40
    POSITIVE LOGITS
     much
    0.68
     few
    0.59
     seldom
    0.56
    yyyy
    0.56
     rarely
    0.56
    much
    0.54
    yyyyy
    0.53
    yyy
    0.52
    thing
    0.51
     możliwe
    0.50
    Activations Density 0.084%

    No Known Activations