© 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. 32500
    Prev
    Next
    INDEX
    Explanations

    whose or which

    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
    onis
    -0.08
     irrespective
    -0.08
    oppins
    -0.08
    angen
    -0.08
     compared
    -0.08
    especially
    -0.08
    uning
    -0.08
    atten
    -0.08
    ener
    -0.08
     perman
    -0.08
    POSITIVE LOGITS
     whose
    0.33
     which
    0.32
    which
    0.30
    whose
    0.29
     WHICH
    0.23
     коÑĤоÑĢ
    0.22
     коÑĤоÑĢÑĭй
    0.22
     wich
    0.21
     whom
    0.19
    (which
    0.18
    Activations Density 0.493%

    No Known Activations