© 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. 34940
    Prev
    Next
    INDEX
    Explanations

    scale, massive scale, commercial scale

    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
    æĬ¹
    -0.10
    æĭį
    -0.09
    ael
    -0.08
    èĵĿ天
    -0.08
    夺
    -0.08
    审åĪ¤
    -0.08
    _instr
    -0.08
    é©¿
    -0.08
    _MISC
    -0.08
    ither
    -0.08
    POSITIVE LOGITS
    è§Ħ模
    0.32
    è¦ı模
    0.25
     scale
    0.21
    大è§Ħ模
    0.20
    è§Ħ模åĮĸ
    0.19
    è§Ħ模æľĢ大
    0.19
    scale
    0.19
    -scale
    0.18
     Scale
    0.17
    (scale
    0.17
    Activations Density 0.076%

    No Known Activations