© 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. 15394
    Prev
    Next
    INDEX
    Explanations

    sentence-ending punctuation followed by capitalized words

    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
    “But
    -0.11
    "But
    -0.11
    uta
    -0.10
    udes
    -0.10
     explan
    -0.10
    uts
    -0.10
     Freak
    -0.09
    VM
    -0.09
     shy
    -0.09
     timid
    -0.09
    POSITIVE LOGITS
     damned
    0.10
    "Yeah
    0.10
     Nah
    0.10
     Nope
    0.10
    åķ
    0.09
     nah
    0.09
     wre
    0.09
    Damn
    0.09
     laz
    0.09
     nodded
    0.09
    Activations Density 0.065%

    No Known Activations