© 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. 35-TRANSCODER-HP
    7. 39680
    Prev
    Next
    INDEX
    Explanations

    -com

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     comb
    -0.34
     conform
    -0.29
    mut
    -0.29
     combo
    -0.28
     conforme
    -0.28
     liqu
    -0.27
     orth
    -0.27
     demonstr
    -0.24
     act
    -0.24
    or
    -0.24
    POSITIVE LOGITS
    b
    0.42
    bat
    0.36
    bach
    0.33
    bv
    0.32
    bau
    0.32
    bate
    0.32
    bite
    0.29
    bilt
    0.29
    bish
    0.28
    bz
    0.28
    Activations Density 0.056%

    No Known Activations