© 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. 30-TRANSCODER-HP
    7. 96570
    Prev
    Next
    INDEX
    Explanations

    Different fish species

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    exas
    -0.29
     каÑĩ
    -0.26
    èĭĴ
    -0.26
    @js
    -0.25
    edList
    -0.24
     FAG
    -0.24
    (sys
    -0.24
    unas
    -0.24
     Fach
    -0.24
     bones
    -0.23
    POSITIVE LOGITS
    bral
    0.27
    .prot
    0.27
    .setColumn
    0.26
    çªĹå¸ĺ
    0.26
    çªĹåı£
    0.26
    bie
    0.25
     Prot
    0.25
    å¦Ĵ
    0.25
    izont
    0.25
    bard
    0.25
    Activations Density 0.886%

    No Known Activations