© 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. 4104
    Prev
    Next
    INDEX
    Explanations

    shots or counts

    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
     flesh
    -0.11
     Conduct
    -0.10
     conduct
    -0.10
     seminar
    -0.10
     conducting
    -0.10
     unr
    -0.09
     choir
    -0.08
    Jac
    -0.08
    дал
    -0.08
     tail
    -0.08
    POSITIVE LOGITS
     shots
    0.26
     strokes
    0.26
     repetitions
    0.23
     iterations
    0.22
     rounds
    0.21
     cycles
    0.20
    shots
    0.20
     Shots
    0.19
     punches
    0.17
     blasts
    0.17
    Activations Density 0.454%

    No Known Activations