© 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. 39222
    Prev
    Next
    INDEX
    Explanations

    "I mean" followed by punctuation or specific 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
    çļĦä¸Ģ款
    -0.11
    chaft
    -0.10
    composite
    -0.09
    å¿ĥ缮
    -0.09
    MEM
    -0.09
    èIJ¥
    -0.09
    çĭĢ
    -0.09
    ä¸ĢéĿ¢
    -0.09
    лав
    -0.09
    çĬ¶
    -0.09
    POSITIVE LOGITS
    estar
    0.10
    è§Į
    0.08
     Boiler
    0.08
    antine
    0.08
    éĹ®éĹ®
    0.08
    个人è§ĤçĤ¹
    0.08
    pent
    0.08
    å®ĥæĺ¯
    0.08
    èĩ
    0.08
    ance
    0.07
    Activations Density 0.006%

    No Known Activations