© 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. 31-TRANSCODER-HP
    7. 18098
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ules
    -0.30
    elly
    -0.27
    elp
    -0.27
    otten
    -0.27
    pa
    -0.26
    elper
    -0.26
    ä¹ĭåĬĽ
    -0.25
     Beats
    -0.25
     Lexer
    -0.24
     upon
    -0.24
    POSITIVE LOGITS
     cro
    0.26
    ç§Ģ丽
    0.26
    群ä½ĵ
    0.25
    天çĦ¶
    0.25
    .cross
    0.25
    è±Ĩçĵ£
    0.25
    _cross
    0.24
    èIJģ
    0.24
    çķĮ
    0.24
    éĴī
    0.23
    Activations Density 0.387%

    No Known Activations