© 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. 7030
    Prev
    Next
    INDEX
    Explanations

    disparage, criticize, or speak negatively

    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
    alar
    -0.10
     bour
    -0.10
     insistence
    -0.09
    ahun
    -0.09
     prompting
    -0.09
     conv
    -0.09
     Urb
    -0.09
     urging
    -0.09
    ä½ľ
    -0.08
     Ze
    -0.08
    POSITIVE LOGITS
     derog
    0.20
     dispar
    0.18
     slander
    0.18
     criticize
    0.17
     criticisms
    0.16
     criticism
    0.15
     negatively
    0.15
     negative
    0.15
     gossip
    0.14
     critic
    0.14
    Activations Density 0.187%

    No Known Activations