© 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. 10616
    Prev
    Next
    INDEX
    Explanations

    defense attorneys arguing cases

    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
    éļ¾åħį
    -0.10
    åĬłå¿«åıijå±ķ
    -0.09
     singleton
    -0.09
    Rows
    -0.09
    venience
    -0.09
    åºij
    -0.09
    åĨķ
    -0.08
    rush
    -0.08
    æĥ©
    -0.08
    entr
    -0.08
    POSITIVE LOGITS
    亲æĥħ
    0.10
    éºĴ
    0.09
    空çĻ½
    0.08
    å¾ĭå¸Ī
    0.08
    串
    0.08
    å®ī举
    0.08
    çĥŁ
    0.08
    亲åıĭ
    0.08
    çľĭä¸Ģçľĭ
    0.08
    èĪŀ
    0.08
    Activations Density 0.053%

    No Known Activations