© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Olmo-3-1125-32B
    5. 32-RES-BATCHTOPK-131K
    7. 17638
    Prev
    Next
    INDEX
    Explanations

    risk and safety concerns

    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
     compulsory
    -0.09
    hor
    -0.09
     mental
    -0.09
     alien
    -0.08
     Alien
    -0.08
     ration
    -0.08
     hor
    -0.08
     vanity
    -0.08
     mentally
    -0.08
    POSITIVE LOGITS
     safest
    0.19
     safer
    0.17
     risk
    0.17
     Risk
    0.16
     risks
    0.16
    Risk
    0.15
    -risk
    0.15
     theoretically
    0.14
     precautions
    0.13
     Transmission
    0.13
    Activations Density 0.232%

    No Known Activations