© 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. 46276
    Prev
    Next
    INDEX
    Explanations

    said followed by quote or clause

    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
     feels
    -0.14
     continues
    -0.12
     has
    -0.12
    —is
    -0.12
     realizes
    -0.12
     appears
    -0.12
     seems
    -0.12
     is
    -0.11
     threatens
    -0.11
     неÑĤ
    -0.11
    POSITIVE LOGITS
     depended
    0.16
     hadn
    0.15
     tended
    0.14
     avait
    0.14
     était
    0.14
     had
    0.14
     existed
    0.13
     could
    0.13
     belonged
    0.13
    had
    0.13
    Activations Density 0.168%

    No Known Activations