© 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. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 60653
    Prev
    Next
    INDEX
    Explanations

    sentences or short phrases that introduce or label the following content (section headers like "Below is", "List of", "Instruction:", etc.).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ToString
    -0.07
     (((
    -0.07
    apas
    -0.07
    打好
    -0.06
     Turns
    -0.06
    /interface
    -0.06
     crochet
    -0.06
    澳大利亚
    -0.06
    _pitch
    -0.06
    -turned
    -0.06
    POSITIVE LOGITS
     Scene
    0.07
    invoke
    0.07
    堕
    0.07
    预先
    0.07
    itol
    0.07
    专属
    0.06
    深切
    0.06
    .advance
    0.06
    parameter
    0.06
    无比
    0.06
    Activations Density 0.055%

    No Known Activations