© 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. 91385
    Prev
    Next
    INDEX
    Explanations

    verbs of speaking followed by 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
     Organization
    -0.09
     backed
    -0.08
     spoken
    -0.08
     overly
    -0.08
     Esper
    -0.08
    éŁ³
    -0.08
     mou
    -0.08
    arro
    -0.08
     pitch
    -0.07
     anomal
    -0.07
    POSITIVE LOGITS
     suddenly
    0.11
    åijĬ
    0.11
     tells
    0.10
     Led
    0.10
     frank
    0.10
     surprised
    0.10
     imaginable
    0.10
    -online
    0.10
     realistically
    0.10
    dq
    0.09
    Activations Density 0.009%

    No Known Activations