© 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. 40095
    Prev
    Next
    INDEX
    Explanations

    1

    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
     low
    -0.33
     nine
    -0.31
    low
    -0.27
     that
    -0.27
     hyp
    -0.27
     possible
    -0.26
     bits
    -0.26
     the
    -0.26
     items
    -0.26
     building
    -0.26
    POSITIVE LOGITS
    seven
    0.28
     Salman
    0.28
    SHA
    0.26
    ASF
    0.26
    SSF
    0.26
    5
    0.26
    VRT
    0.25
     Seventh
    0.25
     ASF
    0.25
     EVP
    0.25
    Activations Density 58.845%

    No Known Activations