© 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. 91801
    Prev
    Next
    INDEX
    Explanations

    staying on track

    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
     buffering
    -0.10
    Mismatch
    -0.09
    ahan
    -0.09
    bits
    -0.09
    oun
    -0.09
    rap
    -0.09
     YT
    -0.08
     Dem
    -0.08
    bare
    -0.08
     storage
    -0.08
    POSITIVE LOGITS
     stays
    0.15
     stay
    0.14
     timely
    0.14
     Stay
    0.13
     stayed
    0.13
    stay
    0.13
     staying
    0.13
     invol
    0.12
     consistently
    0.12
    Stay
    0.12
    Activations Density 0.340%

    No Known Activations