© 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. 22873
    Prev
    Next
    INDEX
    Explanations

    poison, poisoning, poisonous

    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
    è½»æĿ¾
    -0.10
    ÙĪر
    -0.09
    sst
    -0.09
    çĶ¨åĵģ
    -0.09
    umph
    -0.09
    踵
    -0.08
    sie
    -0.08
    ume
    -0.08
    ourn
    -0.08
    å¹Ħ
    -0.08
    POSITIVE LOGITS
    æĻ°
    0.12
    ãģ¾ãģ¾
    0.10
    holes
    0.09
    æ³ī
    0.09
    jabi
    0.09
    ettel
    0.08
    arium
    0.08
    liness
    0.08
    ç¼ķ
    0.08
    esti
    0.08
    Activations Density 0.010%

    No Known Activations