© 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. 6472
    Prev
    Next
    INDEX
    Explanations

    last followed by time period

    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
    ément
    -0.09
    -0.09
    æĭ¾
    -0.08
    éĶĢ
    -0.08
    HELL
    -0.08
    VEN
    -0.08
    çĭĹ
    -0.08
    è§ħ
    -0.08
     instructions
    -0.08
    ocs
    -0.08
    POSITIVE LOGITS
    éĢļè¡Į
    0.11
    eda
    0.09
    éĺĻ
    0.09
    éĥ½æľīèĩªå·±
    0.09
    è¡ĮåĪĹ
    0.09
    å¹´ä¸ĭåįĬå¹´
    0.09
    æ¯ħ
    0.09
    æĺŁè¾°
    0.09
    Ñıм
    0.09
    Checksum
    0.09
    Activations Density 0.015%

    No Known Activations