© 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
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 5923
    Prev
    Next
    INDEX
    Explanations

    Functional connecting words

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĺ²æŃ¢
    -0.47
    éģ¿åħį
    -0.40
    æ¶ĪéĻ¤
    -0.38
    æĬµå¾¡
    -0.37
    æİĴéĻ¤
    -0.36
    对æĬĹ
    -0.35
    以åħį
    -0.35
     prevents
    -0.33
     eliminate
    -0.32
    æīĵåĩ»
    -0.32
    POSITIVE LOGITS
     Stateless
    0.25
     unread
    0.25
    lg
    0.24
    lectic
    0.23
    Ñħодим
    0.23
    没éĴ±
    0.23
    Aqu
    0.23
    .cons
    0.23
    numer
    0.23
    lish
    0.23
    Activations Density 0.066%

    No Known Activations