© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 2-GEMMASCOPE-RES-16K
    11. 15044
    Prev
    Next
    INDEX
    Explanations

    references to linked structures or entities associated with tuples

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     betweenstory
    -0.80
    ulement
    -0.73
    intenant
    -0.59
    berdayakan
    -0.57
    IntoConstraints
    -0.57
     SUSTAIN
    -0.56
    VERTISEMENT
    -0.55
    Obfuscated
    -0.55
     AspNetCore
    -0.54
    Datuak
    -0.54
    POSITIVE LOGITS
     cref
    0.70
    verifyException
    0.67
     instanceof
    0.63
    __(/*!
    0.62
    see
    0.53
    __*/
    0.40
    siehe
    0.39
    modelBuilder
    0.36
    offsetof
    0.35
    Hauptartikel
    0.35
    Activations Density 0.159%

    No Known Activations