© 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. 18774
    Prev
    Next
    INDEX
    Explanations

    scam / fraud

    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
    åĻ´
    -0.10
    çľĭ书
    -0.10
    åħ³éĶ®æŶåĪ»
    -0.10
    顾éĹ®
    -0.09
    erez
    -0.09
    å¹²éĥ¨éĺŁä¼į
    -0.09
    æģ¶åĮĸ
    -0.09
    åĸ·
    -0.09
    读书
    -0.09
    _ant
    -0.09
    POSITIVE LOGITS
    è¯ĪéªĹ
    0.20
    éªĹåŃIJ
    0.20
    éªĹå±Ģ
    0.20
    éªĹ
    0.19
    é¨Ļ
    0.18
    欺è¯Ī
    0.16
     scam
    0.15
     scams
    0.15
    被éªĹ
    0.15
    è©
    0.14
    Activations Density 0.138%

    No Known Activations