© 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. 8282
    Prev
    Next
    INDEX
    Explanations

    vehicles colliding into others

    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
    æĭ¿èµ·
    -0.09
    人éĢī
    -0.09
     Sah
    -0.08
    æ²»çĹħ
    -0.08
    泡泡
    -0.08
    loud
    -0.08
     cr
    -0.08
    ieg
    -0.08
    æĶ¯éħį
    -0.07
    èĸ¤
    -0.07
    POSITIVE LOGITS
    æĴŀ
    0.23
    碰æĴŀ
    0.15
    æĴŀåĩ»
    0.15
     collision
    0.14
    collision
    0.13
     collisions
    0.12
     Collision
    0.11
     accident
    0.11
    糸
    0.11
     collide
    0.11
    Activations Density 0.088%

    No Known Activations