© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 11216
    Prev
    Next
    INDEX
    Explanations

    expressions of unfamiliarity or experiences that have never been encountered before

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

    Prior experience or unfamiliarity

    np_acts-logits-general · gemini-2.0-flash

    never before

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ^(@)
    -0.74
    :+:
    -0.72
    principalColumn
    -0.71
     cherchés
    -0.70
     */
    
    
    -0.67
     Administrativna
    -0.64
    )";
    
    -0.63
     actualidad
    -0.62
     InputDecoration
    -0.62
    ibrill
    -0.61
    POSITIVE LOGITS
     previously
    2.07
     Previously
    1.89
     auparavant
    1.87
     sebelumnya
    1.77
     previous
    1.73
    previously
    1.73
    Previously
    1.70
     ранее
    1.62
     zuvor
    1.56
     PREVIOUS
    1.54
    Activations Density 0.293%

    No Known Activations