© 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. 95372
    Prev
    Next
    INDEX
    Explanations

    peak

    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
    Accepted
    -0.27
    ifers
    -0.26
    ombok
    -0.26
     optimized
    -0.25
    itet
    -0.25
    è¾ij
    -0.25
     expl
    -0.24
    avers
    -0.24
    accepted
    -0.24
    optimized
    -0.24
    POSITIVE LOGITS
    time
    0.45
     time
    0.44
    -time
    0.40
    *time
    0.40
     period
    0.39
     periods
    0.38
     TIME
    0.37
    _period
    0.36
    _time
    0.36
    	time
    0.36
    Activations Density 0.103%

    No Known Activations