© 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. 22244
    Prev
    Next
    INDEX
    Explanations

    factors and variables

    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
    idia
    -0.09
    /assert
    -0.09
    SYM
    -0.08
    _REPEAT
    -0.08
    .ps
    -0.08
    IfNeeded
    -0.08
    æĦıåĽ¾
    -0.08
    èĩ´
    -0.08
    enia
    -0.08
    .pm
    -0.08
    POSITIVE LOGITS
    åĽłç´ł
    0.37
     factors
    0.33
    çļĦåĽłç´ł
    0.31
     Factors
    0.27
     ÑĦакÑĤоÑĢ
    0.22
    Factors
    0.21
    çŃīåĽłç´ł
    0.20
    _factors
    0.20
     factor
    0.18
    éĩįè¦ģåĽłç´ł
    0.18
    Activations Density 0.306%

    No Known Activations