Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 84474
    Prev
    Next
    INDEX
    Explanations

    _replace

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of deprecated PHP regular‐expression functions (e.g. “preg_replace” with the `e` modifier and “eregi”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _delete
    -0.07
    Van
    -0.06
     Merrill
    -0.06
    director
    -0.06
     director
    -0.06
     relating
    -0.06
     ucwords
    -0.06
     değildi
    -0.06
    EINVAL
    -0.06
     VIS
    -0.06
    POSITIVE LOGITS
    ανδ
    0.07
    noxious
    0.06
    ्रत
    0.06
     SCE
    0.06
    biology
    0.06
     Atmospheric
    0.06
    ='',
    0.06
    _pct
    0.06
     Meteor
    0.06
     compt
    0.06
    Activations Density 0.000%

    No Known Activations