© 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. 30650
    Prev
    Next
    INDEX
    Explanations

    codename or name

    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
    å®īå®ģ
    -0.09
    oto
    -0.08
    çĿģå¼Ģ
    -0.08
    9
    -0.08
    8
    -0.08
    åī½
    -0.08
    ä¸į失
    -0.08
    åį«
    -0.08
    .piece
    -0.08
    çĸ²
    -0.08
    POSITIVE LOGITS
    称为
    0.29
    称ä¹ĭ为
    0.29
    åı«åģļ
    0.29
    åIJįåı«
    0.25
    åIJį为
    0.24
     called
    0.23
    稱
    0.22
    ç®Ģ称
    0.22
    称
    0.22
    called
    0.22
    Activations Density 0.266%

    No Known Activations