© 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. 34396
    Prev
    Next
    INDEX
    Explanations

    -ost endings

    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
    ãģ¹ãģį
    -0.14
    erate
    -0.13
    er
    -0.11
    thes
    -0.11
    itty
    -0.11
    pty
    -0.11
     dozen
    -0.10
    ITY
    -0.10
    thic
    -0.10
    halt
    -0.10
    POSITIVE LOGITS
    ively
    0.15
    ech
    0.14
    ion
    0.14
    udio
    0.14
    ype
    0.14
    rength
    0.13
    andard
    0.13
    hetic
    0.13
    aurants
    0.13
    emplate
    0.12
    Activations Density 0.334%

    No Known Activations