© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Olmo-3-1125-32B
    5. 32-RES-BATCHTOPK-131K
    7. 34045
    Prev
    Next
    INDEX
    Explanations

    spatial relationships or joining

    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
    Negative Logits
    onom
    -0.11
    Decoration
    -0.11
    bookmark
    -0.11
     irony
    -0.11
    181
    -0.11
     Rebel
    -0.10
    ray
    -0.10
    ivet
    -0.10
    hani
    -0.10
     barn
    -0.10
    POSITIVE LOGITS
     beside
    0.22
     alongside
    0.19
     opposite
    0.13
     Bes
    0.11
    joining
    0.11
     besides
    0.11
     join
    0.10
     behind
    0.10
     neben
    0.10
     Join
    0.10
    Activations Density 0.081%

    No Known Activations