© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9392
    Prev
    Next
    INDEX
    Explanations

    excuses or blame

    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
     gre
    -0.09
    iram
    -0.09
    æīĵåĩ»
    -0.09
     Harmony
    -0.08
    ferences
    -0.08
    ilon
    -0.08
    nj
    -0.08
     threat
    -0.08
    atically
    -0.08
    ç¬ĥ
    -0.08
    POSITIVE LOGITS
     excuses
    0.15
    解éĩĬ
    0.14
     excuse
    0.13
    辩æĬ¤
    0.13
    辩
    0.13
    åİŁè°ħ
    0.12
    åĢŁåı£
    0.12
     blame
    0.11
    解éĩĭ
    0.11
    èĩªæĪij
    0.11
    Activations Density 0.056%

    No Known Activations