© 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. 14479
    Prev
    Next
    INDEX
    Explanations

    we should

    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
     believe
    -0.16
     believes
    -0.14
     wishes
    -0.12
     Believe
    -0.12
    fon
    -0.12
     wish
    -0.12
     realized
    -0.12
     believed
    -0.11
     feel
    -0.11
    rum
    -0.11
    POSITIVE LOGITS
     should
    0.15
     ought
    0.14
     shouldn
    0.13
     need
    0.12
     could
    0.10
     gotta
    0.10
     Should
    0.10
     needs
    0.10
    should
    0.09
     NEED
    0.09
    Activations Density 0.067%

    No Known Activations