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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 875
    Prev
    Next
    INDEX
    Explanations

    expressions related to physical actions and body parts, particularly involving confrontation or aggression

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

    references to gender dynamics and the comparison of abilities between men and women

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

    ass and buttocks

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    exitRule
    -0.64
    ')['
    -0.63
    expandindo
    -0.59
    awtextra
    -0.59
    دانشنامهٔ
    -0.58
    preventDefault
    -0.57
    KURZBESCHREIBUNG
    -0.57
    principalTable
    -0.57
    DoubleQuotes
    -0.55
    évaluateur
    -0.54
    POSITIVE LOGITS
     butt
    1.39
     ass
    1.36
     buttocks
    1.28
     arse
    1.20
     butts
    1.17
    butt
    1.12
     asses
    1.05
     Butt
    1.00
     ASS
    0.93
    ass
    0.91
    Activations Density 0.272%

    No Known Activations