© 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. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 41671
    Prev
    Next
    INDEX
    Explanations

    recipes and instructions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bre
    -0.27
     Fixed
    -0.26
     wake
    -0.26
    _modify
    -0.25
     wakes
    -0.25
     admissions
    -0.25
     McB
    -0.25
    åĢĴåľ¨åľ°
    -0.24
    ragen
    -0.24
    -uri
    -0.24
    POSITIVE LOGITS
    esium
    0.31
     spr
    0.28
     whole
    0.28
    çIJµ
    0.28
     pl
    0.27
     drawer
    0.26
    odelist
    0.25
    anton
    0.25
    åĽ½éĻħåIJĪä½ľ
    0.24
    assy
    0.24
    Activations Density 0.002%

    No Known Activations