© 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. 15294
    Prev
    Next
    INDEX
    Explanations

    predicting or formatting punctuation

    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
    OTP
    -0.10
    ––
    -0.09
     (=
    -0.08
    otp
    -0.08
     Beverly
    -0.08
    --
    -0.08
     satisfactory
    -0.07
     avenues
    -0.07
    éré
    -0.07
     naï
    -0.07
    POSITIVE LOGITS
     pissed
    0.12
     trag
    0.11
    gabe
    0.11
     nad
    0.11
    Bookmark
    0.10
     backwards
    0.10
     loser
    0.10
     potentially
    0.10
    <article
    0.10
     piss
    0.10
    Activations Density 0.100%

    No Known Activations