© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51248
    Prev
    Next
    INDEX
    Explanations

    cryptography, cipher, algorithm, stream

    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
    ä¸įå¿į
    -0.10
    ond
    -0.09
    å·§
    -0.08
    apel
    -0.08
    å¡
    -0.08
    æ£ļ
    -0.08
    åĵģä½į
    -0.08
     hart
    -0.08
    éľ²å¤©
    -0.08
    4
    -0.07
    POSITIVE LOGITS
    åºĶçĶ¨äºİ
    0.11
    uits
    0.11
    åħŃåįģ
    0.11
    强度
    0.10
    werk
    0.10
    ç¼ij
    0.09
    .opens
    0.09
    èĶŁ
    0.09
    éĥ¨åī¯éĥ¨éķ¿
    0.09
    éĻ·
    0.09
    Activations Density 0.388%

    No Known Activations