© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36937
    Prev
    Next
    INDEX
    Explanations

    negated verbs (wasn't, hadn't, aren't, isn't)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_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
    ãĥ³ãĤ°
    -0.11
    een
    -0.10
    ted
    -0.10
    ters
    -0.10
    ting
    -0.10
     коÑĤоÑĢ
    -0.09
    ts
    -0.09
    ë¡Ń
    -0.09
    pping
    -0.09
    cam
    -0.09
    POSITIVE LOGITS
    etwork
    0.11
    /or
    0.10
    ra
    0.10
    uvo
    0.10
    ally
    0.10
    't
    0.09
    à¹Į
    0.09
    /of
    0.09
    /her
    0.09
    ormal
    0.09
    Activations Density 0.135%

    No Known Activations