© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22916
    Prev
    Next
    INDEX
    Explanations

    predator

    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
    itre
    -0.10
     ÑįкÑģ
    -0.09
    ãģĹãģ¾ãģĨ
    -0.09
     breeze
    -0.09
    ue
    -0.09
    pest
    -0.09
     SOS
    -0.09
    åħ¬å®īå±Ģ
    -0.09
    ull
    -0.08
    uation
    -0.08
    POSITIVE LOGITS
     Ñģобой
    0.11
    ials
    0.11
    istic
    0.10
    ess
    0.10
    etic
    0.10
    èĢħçļĦ
    0.10
    ordial
    0.09
    à¹Į
    0.09
    ively
    0.09
    æ®Ĩ
    0.09
    Activations Density 0.018%

    No Known Activations