© 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. 5658
    Prev
    Next
    INDEX
    Explanations

    doubtful or uncertain quality

    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
    è´¯
    -0.11
    FAILURE
    -0.09
    æŀ¯
    -0.09
    acker
    -0.09
    akes
    -0.09
    失è°ĥ
    -0.08
     lush
    -0.08
    æĶ»åĬ¿
    -0.08
    éĵ¡
    -0.08
    å¤©çľŁ
    -0.08
    POSITIVE LOGITS
     questionable
    0.12
    ä¸įç¡®å®ļ
    0.10
    ä¸ĢéĿ¢
    0.09
     dubious
    0.09
    ä¸įæ¸ħ
    0.09
    è±ļ
    0.09
    å¾®å¦Ļ
    0.08
    æļ§
    0.08
    ochen
    0.08
    ked
    0.08
    Activations Density 0.014%

    No Known Activations