© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 869
    Prev
    Next
    INDEX
    Explanations

    parms postings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    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
    -esque
    -0.38
    Alright
    -0.32
     Alright
    -0.31
     êµīìŀ¥íŀĪ
    -0.31
    fuck
    -0.30
     siden
    -0.30
     subreddit
    -0.28
     alongside
    -0.27
    缸æ¯Ķäºİ
    -0.27
    缸è¾ĥäºİ
    -0.27
    POSITIVE LOGITS
     thru
    0.39
    好象
    0.34
    _parms
    0.33
    Parms
    0.33
    笫
    0.32
    æĬ
    0.32
    çĶ·å£«
    0.29
    éĢĢä¼ij
    0.29
     parms
    0.28
     postings
    0.28
    Activations Density 0.626%

    No Known Activations