© 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. 32387
    Prev
    Next
    INDEX
    Explanations

    content published or 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
    ä¹Ĵä¹ĵçIJĥ
    -0.10
    adoo
    -0.09
    éĩĬ
    -0.09
    tal
    -0.09
    åģļå¾Ĺ
    -0.09
    寨
    -0.09
     Garland
    -0.08
    é¤IJåħ·
    -0.08
    æīĢ以æĪij
    -0.08
    vable
    -0.08
    POSITIVE LOGITS
    WARD
    0.09
    åįģåŃĹ
    0.09
    åĽŀ头
    0.09
    ëĶ©
    0.09
    ied
    0.09
    å¥ĩå¹»
    0.09
    åĺ¶
    0.08
    ç¾İçĻ½
    0.08
    monds
    0.08
    åĵŃ
    0.08
    Activations Density 0.011%

    No Known Activations