© 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. 107949
    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_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
    çķ¥
    -0.29
    antine
    -0.28
    æĹ¶ç©º
    -0.28
    声
    -0.28
    å¯ĨéĽĨ
    -0.28
    äºĶ大
    -0.27
    oo
    -0.27
    ot
    -0.27
    两大
    -0.26
    éĿ¢å¯¹
    -0.26
    POSITIVE LOGITS
    ï¼ģ",
    0.29
    åĤħ
    0.28
    AccessException
    0.28
    abble
    0.26
    lain
    0.26
    ###↵↵
    0.25
    SES
    0.25
    ấu
    0.24
    ...↵↵↵↵↵↵
    0.24
    /pp
    0.24
    Activations Density 0.151%

    No Known Activations