© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 15-GEMMASCOPE-RES-16K
    11. 6336
    Prev
    Next
    INDEX
    Explanations

    references to violence and murder in various contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_15/width_16k/average_l0_131
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    ManagedBean
    -0.55
     كومونز
    -0.50
    RegressionTest
    -0.49
    Życiorys
    -0.47
    delwed
    -0.47
    wapV
    -0.46
    Personendaten
    -0.45
    ("")]
    -0.45
    resave
    -0.45
    CommonModule
    -0.45
    POSITIVE LOGITS
     cruelty
    0.51
     cruel
    0.47
     beginnetje
    0.45
    ⓧ
    0.44
     killing
    0.43
     brutal
    0.41
     justified
    0.41
     ruthless
    0.40
     murder
    0.40
    杀人
    0.40
    Activations Density 0.832%

    No Known Activations