© 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. 35-GEMMASCOPE-RES-16K
    11. 10403
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate interactions and discussions, particularly in the context of education and engagement activities

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     shes
    -0.67
     theyre
    -0.65
     youre
    -0.59
     werewolf
    -0.55
     reggae
    -0.54
     aromatherapy
    -0.54
     gladiator
    -0.53
     yurt
    -0.53
     multiplayer
    -0.53
     hes
    -0.53
    POSITIVE LOGITS
     The
    1.06
     While
    0.95
     It
    0.92
     There
    0.92
     Although
    0.91
     This
    0.91
     When
    0.90
     In
    0.88
     Because
    0.87
     During
    0.86
    Activations Density 3.697%

    No Known Activations