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. 40564
    Prev
    Next
    INDEX
    Explanations

    application

    np_max-act · gemini-2.0-flash

    The neuron activates on words naming court filings or procedural requests (e.g. “application,” “petition,” “motion”).

    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
     Bul
    -0.07
    昇
    -0.07
    徽
    -0.06
     podium
    -0.06
     Cas
    -0.06
    тя
    -0.06
     Huang
    -0.06
     discovered
    -0.06
    Vote
    -0.06
     Banco
    -0.06
    POSITIVE LOGITS
     Dual
    0.06
    InternalEnumerator
    0.06
    meal
    0.06
    нитель
    0.06
    ��
    0.06
     время
    0.06
    -opacity
    0.06
    izer
    0.06
    рава
    0.06
    Enumeration
    0.06
    Activations Density 0.007%

    No Known Activations