© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51394
    Prev
    Next
    INDEX
    Explanations

    deck

    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
    ç¥ł
    -0.09
    asmus
    -0.09
    haul
    -0.09
    éŃĶ
    -0.09
    andal
    -0.08
    ë°°
    -0.08
    ä¸Ģ个个
    -0.08
    cord
    -0.08
    ials
    -0.08
    èĢIJ
    -0.08
    POSITIVE LOGITS
     twin
    0.09
    ä¸Ńéĥ¨
    0.09
    åĨ²åĬ¨
    0.09
    éĥ¨å§Ķ
    0.08
    _FLOW
    0.08
    ÑĪÑĥ
    0.08
    çĮĸ
    0.08
    ä¸ĵéŨ
    0.08
    åĭĥåĭĥ
    0.08
     Osborne
    0.08
    Activations Density 0.027%

    No Known Activations