© 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. 28008
    Prev
    Next
    INDEX
    Explanations

    forcing or compelled actions

    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
    رش
    -0.10
    笼
    -0.09
    reward
    -0.09
    icy
    -0.09
     defaultMessage
    -0.09
    ëħ¼
    -0.08
     embar
    -0.08
    ancy
    -0.08
    åħħåĪĨåĪ©çĶ¨
    -0.08
    æ¶Ī
    -0.08
    POSITIVE LOGITS
    大家
    0.09
    -paced
    0.09
    _into
    0.08
    ONTAL
    0.08
    INLINE
    0.08
    ä¸Ģæŀ¶
    0.08
    lâ
    0.08
    èıĸ
    0.08
    认æ¸ħ
    0.08
    ément
    0.08
    Activations Density 0.030%

    No Known Activations