© 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. 9568
    Prev
    Next
    INDEX
    Explanations

    integrity 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
    -turned
    -0.09
     stake
    -0.09
    æĢ¨
    -0.08
    ols
    -0.08
    å¯Įæ°ij
    -0.08
    SCRIBE
    -0.08
    Ñĩин
    -0.08
    æIJ¬å®¶
    -0.08
     stakes
    -0.08
    åīį
    -0.08
    POSITIVE LOGITS
     integrity
    0.20
    å®Įæķ´æĢ§
    0.19
     Integrity
    0.18
    å°Ĭ严
    0.14
     intact
    0.13
    å®Įæķ´
    0.12
    egrity
    0.11
     inv
    0.10
    平衡
    0.10
     büt
    0.09
    Activations Density 0.050%

    No Known Activations