© 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. 16928
    Prev
    Next
    INDEX
    Explanations

    are followed by permitted or met

    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
     previously
    -0.11
     is
    -0.11
     knowing
    -0.10
    been
    -0.10
     fors
    -0.10
     BEEN
    -0.10
     fearing
    -0.09
     thats
    -0.09
     belonging
    -0.09
    perf
    -0.09
    POSITIVE LOGITS
    nt
    0.36
    /w
    0.29
    /is
    0.22
    ady
    0.22
     indeed
    0.19
     gonna
    0.18
    nts
    0.18
     supposed
    0.17
     meant
    0.16
    NT
    0.15
    Activations Density 0.430%

    No Known Activations