© 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. 30882
    Prev
    Next
    INDEX
    Explanations

    staged a comeback

    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
     greet
    -0.13
     Removal
    -0.12
     permanently
    -0.12
     overlook
    -0.11
     leisure
    -0.11
     dismantle
    -0.11
    oller
    -0.11
     impatient
    -0.11
     Polo
    -0.11
     disabling
    -0.11
    POSITIVE LOGITS
     salvage
    0.29
     regain
    0.27
     redeem
    0.27
     salv
    0.27
     regained
    0.26
     redemption
    0.25
     recovery
    0.25
     redeemed
    0.25
     recover
    0.24
     Recover
    0.24
    Activations Density 0.565%

    No Known Activations