© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Olmo-3-1125-32B
    5. 32-RES-BATCHTOPK-131K
    7. 27658
    Prev
    Next
    INDEX
    Explanations

    making jokes, compliments, or suggestive comments

    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
     dign
    -0.09
    pled
    -0.09
    asia
    -0.09
    象
    -0.08
     schle
    -0.08
     repent
    -0.08
    BBBB
    -0.08
     due
    -0.08
     deceit
    -0.08
    ufen
    -0.08
    POSITIVE LOGITS
     linger
    0.18
     unnecessarily
    0.13
     lingering
    0.13
     suggestive
    0.12
     purposely
    0.12
     subtle
    0.12
     deliberately
    0.12
     lint
    0.12
     compliment
    0.12
     jokes
    0.11
    Activations Density 0.037%

    No Known Activations