© 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. 55234
    Prev
    Next
    INDEX
    Explanations

    mottled, spotted, blotched

    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
    rende
    -0.09
    ä¸įä½ı
    -0.09
    eras
    -0.09
    esthes
    -0.09
    isel
    -0.09
    udios
    -0.08
    åı¦æľī
    -0.08
    èķĻ
    -0.08
    pendicular
    -0.08
    ặ
    -0.08
    POSITIVE LOGITS
     Roth
    0.09
    odus
    0.09
    çĥ©
    0.08
    饰æ¼Ķ
    0.08
    æķ£å¸ĥ
    0.08
    ening
    0.08
    rl
    0.08
    æľ½
    0.08
    ëįĶ
    0.08
    elder
    0.08
    Activations Density 0.023%

    No Known Activations