© 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. 14709
    Prev
    Next
    INDEX
    Explanations

    once in a while

    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
     ALWAYS
    -0.13
     always
    -0.13
     largely
    -0.12
     constantly
    -0.12
     mostly
    -0.12
     mainly
    -0.12
     promptly
    -0.12
     chiefly
    -0.12
    always
    -0.11
    chalk
    -0.11
    POSITIVE LOGITS
     occasionally
    0.16
     someone
    0.16
     Occasionally
    0.14
    someone
    0.13
    ä¼ļ
    0.13
     glimpse
    0.13
     sometimes
    0.13
     Someone
    0.13
     will
    0.12
     Sometimes
    0.12
    Activations Density 0.107%

    No Known Activations