© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 16-GEMMASCOPE-RES-16K
    11. 13333
    Prev
    Next
    INDEX
    Explanations

    references to illicit or dangerous activities and their implications

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

    crime or illegal activities

    np_acts-logits-general · gemini-2.0-flash

    clandestine criminal activity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     oplossing
    -0.48
     vizsg
    -0.47
    balanceOf
    -0.45
     maravill
    -0.45
     auprès
    -0.44
     puits
    -0.43
     colonnes
    -0.43
    数据的
    -0.43
    column
    -0.43
     intervento
    -0.43
    POSITIVE LOGITS
     autorytatywna
    0.87
     @"/
    0.74
    EndContext
    0.72
     cland
    0.72
    +:+
    0.71
     >=",
    0.69
    EDEFAULT
    0.69
    BindingResult
    0.68
     clandestine
    0.68
     undetected
    0.68
    Activations Density 0.625%

    No Known Activations