© 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. 32478
    Prev
    Next
    INDEX
    Explanations

    quit smoking

    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
    haar
    -0.08
    innitus
    -0.08
    æ®Ĥ
    -0.08
    eut
    -0.08
    ibles
    -0.08
    izzato
    -0.08
    ะ
    -0.08
    باء
    -0.08
    жив
    -0.08
    aters
    -0.08
    POSITIVE LOGITS
    çĦ¶æĺ¯
    0.11
    ters
    0.10
    å¾Ĺ好
    0.10
    ï¸
    0.10
    cken
    0.10
    躬
    0.09
    ule
    0.09
    ely
    0.09
    çĤ¹æ»´
    0.09
    ting
    0.09
    Activations Density 0.005%

    No Known Activations