© 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. 62515
    Prev
    Next
    INDEX
    Explanations

    in charge of

    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
    ife
    -0.10
    åīįæīĢ
    -0.09
    æĥłæ°ij
    -0.09
    odal
    -0.08
    HeaderValue
    -0.08
    åĬĽåѦ
    -0.08
    æľĢéĩįè¦ģ
    -0.08
    Ñħож
    -0.08
    semicolon
    -0.08
    rowing
    -0.08
    POSITIVE LOGITS
    mium
    0.10
    gers
    0.10
    auth
    0.09
    mind
    0.09
    aim
    0.09
    urable
    0.08
     liá»ĩt
    0.08
    éĶ·
    0.08
    æĭĪ
    0.08
    ä¸įæĺĵ
    0.08
    Activations Density 0.004%

    No Known Activations