© 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. 48666
    Prev
    Next
    INDEX
    Explanations

    amb- followed by specific suffixes

    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
    ë§Ŀ
    -0.10
    常å§Ķ
    -0.10
    무
    -0.09
    OrUpdate
    -0.09
    arp
    -0.09
    stown
    -0.09
    -know
    -0.09
    opens
    -0.08
    ernels
    -0.08
    EMAIL
    -0.08
    POSITIVE LOGITS
    assador
    0.11
    assadors
    0.11
    ulance
    0.11
    åĭĥåĭĥ
    0.11
     Ñģобой
    0.10
    çĺ´
    0.10
    prit
    0.09
    åĴĮæĶ¯æĮģ
    0.09
    rien
    0.09
    à¸ģาย
    0.08
    Activations Density 0.023%

    No Known Activations