© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 8628
    Prev
    Next
    INDEX
    Explanations

    references to victims and victimization in various contexts

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

    victim and victims

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    𝕫
    -0.50
    fontaine
    -0.50
     Mante
    -0.48
    𝕓
    -0.47
    ขึ้น
    -0.47
    𝕗
    -0.47
     Newell
    -0.47
     Commerce
    -0.46
     Shand
    -0.46
     smelting
    -0.46
    POSITIVE LOGITS
     victim
    1.23
    Victim
    1.20
    victim
    1.18
     Victim
    1.18
     victims
    1.13
     victime
    1.09
     Victims
    1.07
     victimes
    1.02
     vítima
    1.00
     vítimas
    0.93
    Activations Density 0.008%

    No Known Activations