© 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. 22000
    Prev
    Next
    INDEX
    Explanations

    after sequence

    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
    å¸Ľ
    -0.10
    èĶĵ
    -0.09
    urma
    -0.09
    igest
    -0.09
    æľªå¿ħ
    -0.09
    ëĭ´
    -0.09
    ileo
    -0.09
     tongues
    -0.09
     Buccane
    -0.09
    _IMPORTED
    -0.09
    POSITIVE LOGITS
    yna
    0.09
    iph
    0.09
    odi
    0.09
    è·¨
    0.08
    æĹłéļľç¢į
    0.08
    _unused
    0.08
    å®Įå·¥
    0.08
     Hundred
    0.08
    é£ŀè·ĥ
    0.08
    éľ²å¤©
    0.08
    Activations Density 0.134%

    No Known Activations