© 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. 32701
    Prev
    Next
    INDEX
    Explanations

    "would" or "i'd" or "you'd" followed by verb

    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
    otec
    -0.10
     Wolff
    -0.10
     Platt
    -0.10
    iston
    -0.10
    -parse
    -0.10
    ê³
    -0.09
    pedia
    -0.09
     Intro
    -0.09
    TRY
    -0.09
     BI
    -0.09
    POSITIVE LOGITS
     thought
    0.26
     think
    0.22
    thought
    0.20
     Thought
    0.18
    think
    0.18
    Think
    0.18
     Think
    0.17
     THINK
    0.17
     thinks
    0.16
    Thought
    0.15
    Activations Density 0.088%

    No Known Activations