© 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. 12035
    Prev
    Next
    INDEX
    Explanations

    tail parts and control

    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
    åĸī
    -0.10
    ãĥĢãĥ¼
    -0.09
    ä¸Ńéĵģ
    -0.09
    æ§Ľ
    -0.09
    æ¦Ħ
    -0.09
    vara
    -0.09
    车åĨħ
    -0.09
    èµIJ
    -0.09
    车éģĵ
    -0.09
    eced
    -0.08
    POSITIVE LOGITS
    å°¾
    0.19
    尾巴
    0.18
     tail
    0.17
    tail
    0.14
    .tail
    0.14
     Tail
    0.14
     tails
    0.14
    _tail
    0.14
    Tail
    0.13
    TAIL
    0.13
    Activations Density 0.118%

    No Known Activations