© 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. 2211
    Prev
    Next
    INDEX
    Explanations

    cracks and fissures

    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
    寡
    -0.10
    æ·¹
    -0.10
    åĭŁ
    -0.10
    éĤ®å¯Ħ
    -0.09
    计
    -0.09
     counters
    -0.09
    åŀĽ
    -0.09
    è£ķ
    -0.09
    å½ĵåľ°çļĦ
    -0.09
    æľªæĪIJ
    -0.09
    POSITIVE LOGITS
    è£Ĥ
    0.21
     cracks
    0.17
    è£Ĥç¼Ŀ
    0.16
     cracking
    0.15
     cracked
    0.12
     crack
    0.12
    æĸŃè£Ĥ
    0.11
    çļ²
    0.11
     Crack
    0.11
    çł´è£Ĥ
    0.10
    Activations Density 0.041%

    No Known Activations