© 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. 25-TRANSCODER-HP
    7. 108307
    Prev
    Next
    INDEX
    Explanations

    public services

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¿İ
    -0.27
    oms
    -0.26
     afs
    -0.25
     motivations
    -0.25
    club
    -0.24
    dorf
    -0.24
    ãģ©ãģĨ
    -0.24
     tôn
    -0.24
     Nets
    -0.24
     pil
    -0.24
    POSITIVE LOGITS
    代è¨Ģ
    0.32
    ection
    0.28
    Ẹ
    0.25
    好æ¶Īæģ¯
    0.25
    渲
    0.24
    arra
    0.24
    ä½ĵç³»
    0.24
    èľķ
    0.24
    èĬ
    0.24
    iona
    0.24
    Activations Density 0.122%

    No Known Activations