© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Qwen3-4B
    3. 23-TRANSCODER-HP
    4. 162561
    Prev
    Next
    INDEX
    Explanations

    code and technical documentation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hlen
    -0.25
    kdir
    -0.24
    é²ľè¡Ģ
    -0.23
    迹
    -0.23
    eddar
    -0.23
    åĪĿæģĭ
    -0.23
     persist
    -0.23
    Sit
    -0.23
    Dims
    -0.22
    sex
    -0.22
    POSITIVE LOGITS
     https
    0.31
     http
    0.28
    http
    0.28
    {{{
    0.28
    https
    0.27
     Better
    0.27
     Often
    0.26
     Under
    0.26
    :http
    0.25
     Griffith
    0.25
    Activations Density 0.005%

    No Known Activations