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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 11095
    Prev
    Next
    INDEX
    Explanations

    references to advantages and benefits in various contexts

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

    the word "advantages" in comparative or descriptive contexts.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rä
    -0.41
    otto
    -0.35
    willi
    -0.34
    ellis
    -0.34
     Hab
    -0.34
    matches
    -0.33
    lea
    -0.32
     EFF
    -0.32
    bung
    -0.31
     Prin
    -0.31
    POSITIVE LOGITS
     over
    1.36
     над
    1.07
     sobre
    0.99
     asupra
    0.88
    เหน
    0.88
     gegenüber
    0.86
     über
    0.84
     över
    0.84
     boven
    0.82
    over
    0.80
    Activations Density 0.310%

    No Known Activations