© 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. 121379
    Prev
    Next
    INDEX
    Explanations

    miscellaneous text

    np_max-act · gemini-2.0-flash

    tokens that are single letters (especially uppercase single-letter initials in names or headings).

    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
    �
    -0.08
     plea
    -0.07
    萌
    -0.07
    lector
    -0.07
     Vog
    -0.07
    荣幸
    -0.07
    $con
    -0.07
    以上学历
    -0.06
     '<?
    -0.06
     inex
    -0.06
    POSITIVE LOGITS
    惮
    0.08
    参照
    0.07
    游戏装备
    0.07
    upid
    0.07
     paralyzed
    0.06
    Ḻ
    0.06
    ellipsis
    0.06
    𝕷
    0.06
    /fast
    0.06
    BF
    0.06
    Activations Density 1.171%

    No Known Activations