© 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. 49311
    Prev
    Next
    INDEX
    Explanations

    magic spells and casting

    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
    ohl
    -0.11
    Plug
    -0.11
    icie
    -0.10
    outu
    -0.10
    scratch
    -0.10
     Plug
    -0.10
     Satoshi
    -0.10
     Zu
    -0.09
    herit
    -0.09
    Uploaded
    -0.09
    POSITIVE LOGITS
     spell
    0.29
     spells
    0.29
     magic
    0.25
     conj
    0.23
     magically
    0.21
     enchant
    0.21
     magical
    0.19
     glamour
    0.19
     enchanted
    0.18
     Conj
    0.17
    Activations Density 0.284%

    No Known Activations