© 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. 15639
    Prev
    Next
    INDEX
    Explanations

    usually or typically followed by verb

    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
     increasingly
    -0.12
     fortunately
    -0.11
     presently
    -0.11
     conveniently
    -0.10
     bols
    -0.10
     fy
    -0.10
    ynes
    -0.10
     ther
    -0.10
     enclosing
    -0.10
     Surely
    -0.10
    POSITIVE LOGITS
     usually
    0.25
     typically
    0.21
     Usually
    0.21
     often
    0.20
    usually
    0.20
     Typically
    0.19
    Usually
    0.18
     Often
    0.17
     may
    0.16
    Often
    0.16
    Activations Density 0.204%

    No Known Activations