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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Olmo-3-1125-32B
    5. 32-RES-BATCHTOPK-131K
    7. 89817
    Prev
    Next
    INDEX
    Explanations

    killed, attacked, targeting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    bcywinski/Olmo-3-32B-Base-SAE/saes_allenai_Olmo-3-1125-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     secretly
    -0.11
     undercover
    -0.11
     brav
    -0.10
     covert
    -0.10
     recalls
    -0.10
     Intervention
    -0.10
     begging
    -0.10
     evangel
    -0.09
     joins
    -0.09
    .flex
    -0.09
    POSITIVE LOGITS
     targeting
    0.16
     attacking
    0.13
     attack
    0.13
     target
    0.12
     murdering
    0.12
    targets
    0.12
     targets
    0.12
     kill
    0.12
     succeed
    0.11
     kills
    0.11
    Activations Density 0.276%

    No Known Activations