© 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. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 831
    Prev
    Next
    INDEX
    Explanations

    attack action, penalty, scored

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    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
    äºĴ缸
    -0.27
    缸èģļ
    -0.25
    atitude
    -0.24
     temper
    -0.24
    ime
    -0.23
    éģĩè§ģ
    -0.23
    缸éģĩ
    -0.23
    åħĪ
    -0.23
    äºĴ
    -0.23
    å½¼æѤ
    -0.22
    POSITIVE LOGITS
    æĥ©ç½ļ
    0.28
    éĹ´æİ¥
    0.28
    Immediate
    0.27
     ensuing
    0.27
    åıįèĢĮ
    0.26
    åıªå¥½
    0.26
    åıįåĩ»
    0.26
     penalty
    0.25
    ç½ļ
    0.25
     retal
    0.24
    Activations Density 0.014%

    No Known Activations