© 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. 46893
    Prev
    Next
    INDEX
    Explanations

    repeated requests or refusal

    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
     unintended
    -0.11
     Dek
    -0.10
    apas
    -0.10
     Dank
    -0.10
    advert
    -0.10
    rong
    -0.09
     underrated
    -0.09
     gens
    -0.09
    Cong
    -0.09
     fal
    -0.09
    POSITIVE LOGITS
     consistently
    0.13
     incess
    0.12
     refusal
    0.12
     constantly
    0.12
     behavior
    0.11
     refused
    0.11
     requests
    0.11
     everytime
    0.11
     attempts
    0.11
     anytime
    0.11
    Activations Density 0.059%

    No Known Activations