© 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. 28157
    Prev
    Next
    INDEX
    Explanations

    equilibrium state

    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¢ŀèµĦ
    -0.10
    沫
    -0.10
    ç¿»
    -0.09
    инд
    -0.09
    æĥ
    -0.09
    .Param
    -0.08
    buff
    -0.08
    ä¼łåªĴ
    -0.08
    arine
    -0.08
    greater
    -0.08
    POSITIVE LOGITS
     equilibrium
    0.15
    平衡
    0.14
    稳å®ļ
    0.12
    çĬ¶æĢģä¸ĭ
    0.12
    稳å®ļçļĦ
    0.11
    åĿĩè¡¡
    0.11
    ilibrium
    0.11
    å®īå®ļ
    0.11
    çĬ¶æĢģ
    0.10
    稳
    0.10
    Activations Density 0.171%

    No Known Activations