© 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. 25-TRANSCODER-HP
    7. 107384
    Prev
    Next
    INDEX
    Explanations

    from

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    鸶
    -0.27
    ijke
    -0.26
     pressures
    -0.26
    gren
    -0.25
    åıĸæ¶Ī
    -0.25
     diaper
    -0.25
    idine
    -0.25
    idders
    -0.24
    å¦ĥ
    -0.24
    ,temp
    -0.24
    POSITIVE LOGITS
     http
    0.32
    cast
    0.30
    http
    0.27
     G
    0.27
     https
    0.27
    æīĭåĨĮ
    0.26
    -na
    0.26
     paste
    0.26
    ynam
    0.26
     C
    0.26
    Activations Density 0.918%

    No Known Activations