© 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. 33023
    Prev
    Next
    INDEX
    Explanations

    topics and their states

    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
     Means
    -0.11
    means
    -0.10
     means
    -0.09
     Saw
    -0.08
    says
    -0.08
     legitim
    -0.08
     whereas
    -0.08
     besides
    -0.07
     believing
    -0.07
    imates
    -0.07
    POSITIVE LOGITS
     undergo
    0.17
     become
    0.14
     becomes
    0.14
     magically
    0.14
     appears
    0.13
     remained
    0.13
     linger
    0.13
     gradually
    0.13
     briefly
    0.12
     disappears
    0.12
    Activations Density 1.047%

    No Known Activations