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. 555
    Prev
    Next
    INDEX
    Explanations

    keywords related to discrimination and historical injustices

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

    Discussions of slavery and racial issues

    np_acts-logits-general · gemini-2.0-flash

    slavery and racial segregation

    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
    featureID
    -0.70
     poisson
    -0.55
    ByVersion
    -0.51
     Bore
    -0.50
     punct
    -0.48
    IFS
    -0.48
    enterOuterAlt
    -0.47
    umont
    -0.45
     hacker
    -0.45
    fortawesome
    -0.45
    POSITIVE LOGITS
     racial
    1.45
     racism
    1.43
     racist
    1.40
     racially
    1.31
     Racial
    1.25
     Racism
    1.22
    racial
    1.20
     segregation
    1.17
     race
    1.15
    Racism
    1.13
    Activations Density 0.543%

    No Known Activations