© 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. 48079
    Prev
    Next
    INDEX
    Explanations

    individualized

    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
    ndx
    -0.10
    ÙĪØ·
    -0.09
    è¿ĩçļĦ
    -0.09
    forced
    -0.09
    cre
    -0.09
    /dc
    -0.09
    ç¼ģ
    -0.09
    isode
    -0.09
    ryo
    -0.09
    åħħåĪĨåĪ©çĶ¨
    -0.08
    POSITIVE LOGITS
    ity
    0.17
    ized
    0.16
    mente
    0.15
    istic
    0.14
    ize
    0.13
    izing
    0.12
    izes
    0.11
    ised
    0.11
    iating
    0.11
    ially
    0.11
    Activations Density 0.024%

    No Known Activations