© 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. Gemma-2-2B
    5. 2-RES-MATRYOSHKA-DC
    7. 6465
    Prev
    Next
    INDEX
    Explanations

    expressions of negative past tense prohibitions or denials

    oai_token-act-pair · gpt-4.1-nano
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    don
    -1.55
     Don
    -1.49
    Don
    -1.38
     don
    -1.37
     DON
    -1.30
    DON
    -1.27
    dont
    -1.23
     Dont
    -1.13
    Dont
    -1.11
     dont
    -1.06
    POSITIVE LOGITS
     wasn
    1.29
     isn
    1.27
    wasn
    1.23
     weren
    1.16
     Wasn
    1.16
     didn
    1.05
     hasn
    1.05
    Wasn
    1.02
    isn
    1.00
    Isn
    0.99
    Activations Density 0.070%

    No Known Activations