© 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. 26936
    Prev
    Next
    INDEX
    Explanations

    relative pronoun "which"

    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
    ault
    -0.11
     ведÑĮ
    -0.08
    uly
    -0.08
    4
    -0.08
    å··
    -0.08
    äºĽä»Ģä¹Ī
    -0.08
    æĢİä¹Īçľĭ
    -0.08
    è·Ħ
    -0.08
    5
    -0.08
    Opens
    -0.08
    POSITIVE LOGITS
    erner
    0.10
    imal
    0.09
    _prep
    0.09
    ãĤ¤ãĥ«
    0.08
    âĮł
    0.08
    opi
    0.08
    ãĥ³
    0.08
    Playable
    0.08
    ron
    0.08
    born
    0.08
    Activations Density 0.147%

    No Known Activations