© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37683
    Prev
    Next
    INDEX
    Explanations

    suggest that

    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
    anche
    -0.09
    alus
    -0.09
    -scale
    -0.09
    itness
    -0.09
    iets
    -0.09
    ä¿Ĺè¯Ŀ说
    -0.08
    lav
    -0.08
    æĿ³
    -0.08
    æĥľ
    -0.08
    åĬĽæ±Ĥ
    -0.08
    POSITIVE LOGITS
    ively
    0.15
    ably
    0.13
    ors
    0.11
     Ñģобой
    0.11
     rằng
    0.10
    entially
    0.10
    æŃ¤äºº
    0.09
    ivic
    0.09
    大家éĥ½åľ¨
    0.09
    apore
    0.09
    Activations Density 0.032%

    No Known Activations