© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 5510
    Prev
    Next
    INDEX
    Explanations

    references to rape and associated accusations or legal implications

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

    rape and sexual assault

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects mentions of sexual violence and closely related concepts (perpetration, accusations, victims, and services/referrals).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Abit
    -0.40
    MergeFrom
    -0.40
     bezeichneter
    -0.38
    __':
    
    -0.37
    sidemargin
    -0.35
     GoogleFonts
    -0.35
     useStyles
    -0.33
    Groeten
    -0.33
    MENAFN
    -0.33
     doInBackground
    -0.33
    POSITIVE LOGITS
    setVerticalGroup
    0.75
     rape
    0.69
    Rape
    0.60
     Rape
    0.60
     raped
    0.55
    rape
    0.54
     raping
    0.48
    دانشنامهٔ
    0.47
     sexual
    0.47
     ویکی‌پدی
    0.47
    Activations Density 0.201%

    No Known Activations