© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21535
    Prev
    Next
    INDEX
    Explanations

    fine, finer, finest

    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
    åįģä¸ĩ
    -0.09
    æĺŃ
    -0.09
    èĻİ
    -0.09
    åIJīåĪ©
    -0.09
    è£Ļ
    -0.09
    éĺĶ
    -0.09
    çİ©æ³ķ
    -0.08
    outh
    -0.08
    aha
    -0.08
    overs
    -0.08
    POSITIVE LOGITS
    ç»Ĩ
    0.17
    Fine
    0.14
    fine
    0.14
     Fine
    0.14
     fine
    0.14
     finer
    0.11
    FINE
    0.11
    ç»Ĩèħ»
    0.10
    ç»Ĩèĩ´
    0.10
     fines
    0.10
    Activations Density 0.030%

    No Known Activations