© 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. 9066
    Prev
    Next
    INDEX
    Explanations

    bummed or ham words

    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
    oom
    -0.10
    Ĥæķ°
    -0.09
    æīŃæĽ²
    -0.09
    ives
    -0.08
    IVE
    -0.08
    迦
    -0.08
    -spe
    -0.08
    æĸ©
    -0.08
    ories
    -0.08
    ician
    -0.08
    POSITIVE LOGITS
    å¸ĪåħĦ
    0.08
     bum
    0.08
    é¢ı
    0.08
    reau
    0.08
    /student
    0.08
    äºĨ个
    0.08
    ifest
    0.08
    sWith
    0.08
     Soap
    0.07
    iest
    0.07
    Activations Density 0.005%

    No Known Activations