© 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. 15337
    Prev
    Next
    INDEX
    Explanations

    two of them

    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
     gratuit
    -0.10
     franch
    -0.08
     tras
    -0.08
     Fireplace
    -0.08
    isl
    -0.08
     contin
    -0.08
    á»ī
    -0.08
    oup
    -0.08
     disb
    -0.07
     grat
    -0.07
    POSITIVE LOGITS
     who
    0.21
     whom
    0.14
    who
    0.13
     whose
    0.13
     others
    0.12
     men
    0.12
    —who
    0.12
     Others
    0.12
     individuals
    0.11
     souls
    0.11
    Activations Density 0.103%

    No Known Activations