© 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. 37879
    Prev
    Next
    INDEX
    Explanations

    sports positions

    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
    å½ĵäºĭ
    -0.10
    4
    -0.09
    6
    -0.09
    åįĹæŀģ
    -0.09
    趸
    -0.09
    7
    -0.08
    离ä¸įå¼Ģ
    -0.08
    æĪIJè¯Ń
    -0.08
    5
    -0.08
    éĻ²
    -0.08
    POSITIVE LOGITS
    åħ¼
    0.11
    å¤Ħ
    0.09
    èģª
    0.09
    ilder
    0.09
     billig
    0.09
    dont
    0.08
    archy
    0.08
    å°¸
    0.08
    radient
    0.08
    ой
    0.08
    Activations Density 0.007%

    No Known Activations