© 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. 3811
    Prev
    Next
    INDEX
    Explanations

    would followed by a 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
     unf
    -0.13
     recap
    -0.11
    ecc
    -0.11
     multis
    -0.11
    isser
    -0.11
    essel
    -0.10
    lest
    -0.10
     triumph
    -0.10
    ertext
    -0.10
    holm
    -0.10
    POSITIVE LOGITS
     need
    0.13
     likely
    0.11
     Need
    0.10
     tend
    0.09
    allon
    0.09
    need
    0.09
     barr
    0.09
     depend
    0.09
     unlikely
    0.08
    Need
    0.08
    Activations Density 0.184%

    No Known Activations