© 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. 23093
    Prev
    Next
    INDEX
    Explanations

    from the use of

    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
    æ¸ħåįİ大åѦ
    -0.10
    æµļ
    -0.09
    è¡·
    -0.09
    otec
    -0.09
    夹
    -0.09
    å¤
    -0.09
    ãģ£ãģįãĤĬ
    -0.08
    è¿ĽäºĨ
    -0.08
    éĺ²æݧ
    -0.08
    ä»Ģä¹Īä¸ľè¥¿
    -0.08
    POSITIVE LOGITS
     Creed
    0.09
    CTX
    0.09
    uite
    0.09
    åĪĽ
    0.08
    elig
    0.08
    tam
    0.08
    顽
    0.08
    æĻ¾
    0.08
    Codec
    0.08
    ÑĤик
    0.08
    Activations Density 0.011%

    No Known Activations