© 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. 58482
    Prev
    Next
    INDEX
    Explanations

    open land, plains, prairie fire

    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
    ç¦ħ
    -0.09
     cann
    -0.08
    alis
    -0.08
    侨
    -0.08
     Off
    -0.08
    oked
    -0.08
    .Mode
    -0.08
    ;base
    -0.08
    ybrid
    -0.08
    çŃĸåĪĴ
    -0.07
    POSITIVE LOGITS
    å¹³åĿ¦
    0.18
    å¹³åİŁ
    0.17
    èįīåİŁ
    0.13
    æĹ·
    0.13
    å¼ĢéĺĶ
    0.13
    广éĺĶ
    0.13
    å¼ĢæľĹ
    0.12
     plains
    0.12
    çĶ°éĩİ
    0.12
    çī§åľº
    0.12
    Activations Density 0.077%

    No Known Activations