© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21184
    Prev
    Next
    INDEX
    Explanations

    way followed by direction

    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
     Ñģвое
    -0.10
    åĪ«çļĦ
    -0.09
    没äºĭ
    -0.09
    åħ¶
    -0.09
    ông
    -0.09
    ä»ĸèĩªå·±
    -0.09
    èµ°åĩºåİ»
    -0.09
    itas
    -0.08
    åĪ«äºº
    -0.08
    junction
    -0.08
    POSITIVE LOGITS
    zelf
    0.10
    ëģĶ
    0.09
    idl
    0.09
     Ventura
    0.09
    å¥Ħ
    0.08
    íĮħ
    0.08
    tle
    0.08
    APO
    0.08
    åŁºæľ¬æĥħåĨµ
    0.08
    orate
    0.08
    Activations Density 0.005%

    No Known Activations