© 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. 1324
    Prev
    Next
    INDEX
    Explanations

    thorough understanding or care

    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
    eler
    -0.11
    羣æŃ£åģļåĪ°
    -0.09
    ihil
    -0.09
    /rem
    -0.09
    upported
    -0.09
    ularity
    -0.09
    igel
    -0.08
    è´¸
    -0.08
    ÑĥÑĩ
    -0.08
    ually
    -0.08
    POSITIVE LOGITS
    bred
    0.18
    ness
    0.14
    going
    0.14
    fare
    0.11
    åľ°è¯´
    0.10
    NESS
    0.10
    ment
    0.10
    ly
    0.10
    er
    0.09
    æĸ¯åĿ¦
    0.09
    Activations Density 0.009%

    No Known Activations