© 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. 7965
    Prev
    Next
    INDEX
    Explanations

    understanding, knowing, intuit

    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
     plutôt
    -0.12
     understandably
    -0.11
     seriously
    -0.11
    until
    -0.11
    WS
    -0.11
    Until
    -0.11
    azi
    -0.11
    ->[
    -0.11
    SE
    -0.10
    -reply
    -0.10
    POSITIVE LOGITS
     thanks
    0.14
     intimately
    0.14
     knew
    0.11
     firsthand
    0.10
     vaguely
    0.09
     courtesy
    0.09
     already
    0.09
     better
    0.09
     instinct
    0.09
     intuit
    0.08
    Activations Density 0.080%

    No Known Activations