© 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. 54669
    Prev
    Next
    INDEX
    Explanations

    almost never or almost always

    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
    æ£ĭ
    -0.09
    cco
    -0.09
    haft
    -0.09
    dest
    -0.09
    ilian
    -0.09
    hai
    -0.08
    oplast
    -0.08
    ç³ĸ
    -0.08
    æ²Ľ
    -0.08
    çĸı导
    -0.08
    POSITIVE LOGITS
    åĩłä¹İæīĢæľī
    0.11
    æīĢæľīçļĦ
    0.10
    æīĢæľī人éĥ½
    0.09
    .AutoScaleMode
    0.09
    ridge
    0.09
    äºĮ次
    0.09
     jede
    0.09
    éĤ£æ¬¡
    0.09
    ovie
    0.09
    gary
    0.09
    Activations Density 0.014%

    No Known Activations