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. 16202
    Prev
    Next
    INDEX
    Explanations

    questions and constraints

    np_max-act · gemini-2.0-flash

    This neuron detects prescriptive advice phrasing, particularly the pattern “it is important to ….”

    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
     BBQ
    -0.06
     productive
    -0.06
     tanggal
    -0.06
    ніп
    -0.06
    ints
    -0.06
     sideline
    -0.06
    kur
    -0.06
    won
    -0.06
    angered
    -0.06
    ucing
    -0.06
    POSITIVE LOGITS
    "Not
    0.07
     нагруз
    0.07
    _assignment
    0.06
     Cardio
    0.06
    _AD
    0.06
    Not
    0.06
     되었
    0.06
    (nb
    0.06
     warto
    0.06
     처
    0.06
    Activations Density 0.074%

    No Known Activations