© 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. 32233
    Prev
    Next
    INDEX
    Explanations

    your skills are a challenge

    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
    详ç»Ĩä»ĭç»į
    -0.09
    ä¾¿å®ľ
    -0.09
    çĿģå¼Ģ
    -0.09
    åıijå±ķæł¼å±Ģ
    -0.09
    æĪij们è¦ģ
    -0.09
    åĿıäºĭ
    -0.08
    éĽĨåĽ¢èĤ¡ä»½
    -0.08
     gấp
    -0.08
    æĦŁæħ¨
    -0.08
    çĴ
    -0.08
    POSITIVE LOGITS
    ç³»
    0.10
    ties
    0.09
    matches
    0.09
    èĦī
    0.09
     matches
    0.09
     дек
    0.09
    .Match
    0.09
    èĩªæĦ¿
    0.09
    å¯Ħ
    0.08
    .matches
    0.08
    Activations Density 0.052%

    No Known Activations