© 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. 39756
    Prev
    Next
    INDEX
    Explanations

    say "TH"

    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
     auditing
    -0.25
     estimating
    -0.24
     advent
    -0.24
    é¾Ħ
    -0.24
     politely
    -0.24
    èª¿æŁ¥
    -0.24
    -lined
    -0.23
     invent
    -0.23
    -budget
    -0.23
     hurdle
    -0.23
    POSITIVE LOGITS
    un
    0.30
    ol
    0.29
    ron
    0.29
    ure
    0.28
    oke
    0.28
    åį³
    0.27
    ran
    0.27
    oken
    0.26
    oko
    0.25
    ail
    0.25
    Activations Density 0.024%

    No Known Activations