© 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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13230
    Prev
    Next
    INDEX
    Explanations

    legal terms and references to criminal charges and convictions

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

    Legal charges and crimes

    np_acts-logits-general · gemini-2.0-flash

    criminal acts and offenses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    EndInit
    -0.71
    GraphicsUnit
    -0.65
     pienso
    -0.59
    +#+#
    -0.58
    enumi
    -0.57
    出版年
    -0.55
    TypedDataSet
    -0.51
     disfru
    -0.51
     himo
    -0.50
    reel
    -0.48
    POSITIVE LOGITS
     breach
    0.69
     Breach
    0.65
     attempted
    0.64
     murder
    0.57
     possession
    0.56
     assault
    0.55
     negligent
    0.54
    raborty
    0.54
    لاثة
    0.54
     breaching
    0.53
    Activations Density 0.587%

    No Known Activations