© 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 EvalsExportsAPI 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. 77528
    Prev
    Next
    INDEX
    Explanations

    the starts of chemical compound names (the token at the beginning of a compound/drug/chemical term, often in "Applications of X" prompts).

    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
     tones
    -0.07
    عنا
    -0.07
     rebound
    -0.07
     Turner
    -0.07
    صرف
    -0.07
     elementos
    -0.07
     energ
    -0.07
    idente
    -0.07
    存款
    -0.07
    caret
    -0.07
    POSITIVE LOGITS
    .between
    0.08
     androidx
    0.08
    squ
    0.07
     SIX
    0.07
     Along
    0.07
    fé
    0.07
    㟃
    0.07
    ﭩ
    0.07
    (times
    0.07
    얏
    0.07
    Activations Density 0.437%

    No Known Activations