© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 129207
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¸įäºĪ
    -0.46
    æĹłæķĪ
    -0.42
    å°±ä¸įèĥ½
    -0.41
    ä¸įéĢĤåIJĪ
    -0.40
    éĥ½ä¸įèĥ½
    -0.40
    æĭĴç»Ŀ
    -0.39
    ä¸įè¡Į
    -0.39
    æĭĴ
    -0.37
     ineffective
    -0.36
     nowhere
    -0.35
    POSITIVE LOGITS
    顺çķħ
    0.27
    éħIJ
    0.27
    coles
    0.26
     dau
    0.26
    tro
    0.26
    æĶ¾å¿ĥ
    0.25
    ysis
    0.24
     replacements
    0.24
     willing
    0.24
    çĪ·
    0.24
    Activations Density 0.001%

    No Known Activations