© 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. 91660
    Prev
    Next
    INDEX
    Explanations

    can/could followed by sensory 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
    een
    -0.10
     uni
    -0.10
     preferably
    -0.08
     gim
    -0.08
     together
    -0.08
     succ
    -0.08
    erry
    -0.08
     Sem
    -0.08
    ez
    -0.08
     w
    -0.08
    POSITIVE LOGITS
     overhe
    0.14
     hear
    0.14
    ’t
    0.13
    't
    0.13
     heard
    0.11
     immedi
    0.11
    nt
    0.11
     see
    0.11
     hears
    0.11
     only
    0.10
    Activations Density 0.007%

    No Known Activations