© 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. 15004
    Prev
    Next
    INDEX
    Explanations

    balcony, terrace, deck, railing

    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
     Underground
    -0.13
     underground
    -0.12
     buried
    -0.12
     dug
    -0.11
    Locker
    -0.10
     swamp
    -0.10
     unearth
    -0.10
     bury
    -0.10
     crust
    -0.09
     Village
    -0.09
    POSITIVE LOGITS
     railing
    0.24
     balcony
    0.24
     balcon
    0.19
     terrace
    0.18
     Deck
    0.17
     ledge
    0.17
    Deck
    0.16
     rooftop
    0.15
     overlooking
    0.15
     deck
    0.15
    Activations Density 0.176%

    No Known Activations