Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 8739
    Prev
    Next
    INDEX
    Explanations

    terms related to prison, war, and connotation of confinement or capture

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

    preceding "camp" or "facilities"

    np_acts-logits-general · gemini-2.0-flash

    concentration camps and detention

    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
     виправивши
    -0.70
     &___
    -0.63
    BrowserModule
    -0.60
    تقاوى
    -0.59
    RegistryLite
    -0.56
    UIControlState
    -0.54
     typelib
    -0.54
    elemField
    -0.51
    CodeAttribute
    -0.51
    ConstraintMaker
    -0.51
    POSITIVE LOGITS
     facility
    0.92
     boarding
    0.86
     facilities
    0.81
    facility
    0.79
     camp
    0.79
     institution
    0.78
     institutions
    0.75
     nursing
    0.75
     institutional
    0.72
     Boarding
    0.72
    Activations Density 0.185%

    No Known Activations