© 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. Gemma-2-2B
    5. 12-RES-MATRYOSHKA-DC
    7. 14829
    Prev
    Next
    INDEX
    Explanations

    references to harm in various contexts

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WebElementEntity
    -0.78
     Leighton
    -0.77
    Tach
    -0.77
    \}\\
    -0.74
     Dowling
    -0.73
     gynhyrchwyd
    -0.71
    例文帳に追加
    -0.71
    element
    -0.70
     Dresden
    -0.69
    "]);
    -0.69
    POSITIVE LOGITS
     harm
    1.23
    harm
    1.23
     Harm
    1.17
    Harm
    1.16
     harms
    1.11
     Harms
    1.09
    Harmful
    1.05
     Hurt
    1.02
     harmed
    0.91
    Hurt
    0.88
    Activations Density 0.008%

    No Known Activations