© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 8898
    Prev
    Next
    INDEX
    Explanations

    obituaries

    np_max-act · gemini-2.0-flash

    The neuron fires on the introductory descriptive phrases after a person’s name—particularly the “a/whose/who …” appositive clauses that state their role or notable attribute.

    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
    744
    -0.06
     Advertisement
    -0.06
     atheist
    -0.06
     outrageous
    -0.06
     {?}
    -0.06
    eff
    -0.06
    _DATABASE
    -0.06
    049
    -0.06
    stdint
    -0.06
     faithful
    -0.06
    POSITIVE LOGITS
     startPoint
    0.07
    之
    0.06
     consequently
    0.06
    idan
    0.06
     thrown
    0.06
    ERING
    0.06
    pler
    0.06
     reordered
    0.06
    )↵↵↵↵↵↵
    0.06
    曲
    0.06
    Activations Density 0.030%

    No Known Activations