© 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. 58108
    Prev
    Next
    INDEX
    Explanations

    taste and aroma descriptions

    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
     trib
    -0.08
    icity
    -0.08
    çĥŁçģ«
    -0.08
    di
    -0.07
    ç³»ç»Ł
    -0.07
     Dip
    -0.07
    asions
    -0.07
    Ted
    -0.07
    owo
    -0.07
    dio
    -0.07
    POSITIVE LOGITS
    edException
    0.10
    åºĦåĽŃ
    0.09
    azzi
    0.09
    è¿ĻçĤ¹
    0.09
    _EDGE
    0.09
    bih
    0.09
    ican
    0.09
    تخ
    0.09
    bones
    0.09
    èĹIJ
    0.08
    Activations Density 0.001%

    No Known Activations