© 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. 1233
    Prev
    Next
    INDEX
    Explanations

    in source or compliance

    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
    æĴĴ
    -0.08
     lis
    -0.08
    æ¯ķ
    -0.08
    ç£ģ
    -0.08
    IGH
    -0.08
    ç«¥
    -0.08
    ÙĪÙĬÙĦ
    -0.08
    _specific
    -0.08
     illustr
    -0.08
    åıijè¾¾åĽ½å®¶
    -0.08
    POSITIVE LOGITS
    åı£å¾Ħ
    0.10
    平淡
    0.09
    åı£æ°Ķ
    0.09
    или
    0.09
    èĬĥ
    0.08
    uben
    0.08
    aders
    0.08
    ìļķ
    0.08
    upal
    0.08
    åįī
    0.08
    Activations Density 0.032%

    No Known Activations