© 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. 91608
    Prev
    Next
    INDEX
    Explanations

    fake or dishonest statements

    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
    Restr
    -0.10
     Restrictions
    -0.10
     restrictions
    -0.10
     attract
    -0.10
     restriction
    -0.10
     attracting
    -0.09
     appealed
    -0.09
     vet
    -0.09
    applicant
    -0.09
     unfore
    -0.09
    POSITIVE LOGITS
     fake
    0.22
     Fake
    0.21
     faker
    0.19
    fake
    0.18
    Fake
    0.17
     dishonest
    0.16
     silently
    0.16
     fals
    0.16
     falsely
    0.14
     pretending
    0.14
    Activations Density 0.011%

    No Known Activations