© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Olmo-3-1125-32B
    5. 32-RES-BATCHTOPK-131K
    7. 50843
    Prev
    Next
    INDEX
    Explanations

    phrases ending with commas

    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
    Negative Logits
     especially
    -0.13
    especially
    -0.12
     Especially
    -0.10
     Brick
    -0.10
     particularly
    -0.09
     especialmente
    -0.09
    urai
    -0.08
     scor
    -0.08
    atus
    -0.08
     cher
    -0.08
    POSITIVE LOGITS
     meaning
    0.15
     effectively
    0.13
     except
    0.13
     unless
    0.12
    unless
    0.12
    via
    0.12
    meaning
    0.12
     Meaning
    0.11
    boo
    0.11
     ostensibly
    0.11
    Activations Density 0.211%

    No Known Activations