© 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. 5272
    Prev
    Next
    INDEX
    Explanations

    world record, best, holder

    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
     unprecedented
    -0.20
     unparalleled
    -0.12
     rare
    -0.11
     rarity
    -0.11
     Winner
    -0.11
    ìĥī
    -0.10
     exceptional
    -0.10
    precedented
    -0.10
     groundbreaking
    -0.10
    orst
    -0.09
    POSITIVE LOGITS
    holds
    0.14
     Records
    0.13
     holder
    0.13
     holds
    0.12
     Guill
    0.12
     hold
    0.12
     Hold
    0.12
     holding
    0.12
     records
    0.12
     ratified
    0.12
    Activations Density 0.160%

    No Known Activations