© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 44285
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on the placeholder name tokens (e.g. “NAME_1,” “NAME_2,” etc.), i.e. it detects mentions of the characters by their NAME_# identifiers.

    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
    ‌دهد
    -0.06
    istance
    -0.06
    さら
    -0.06
    ниці
    -0.06
    _pixel
    -0.06
    strcasecmp
    -0.06
     başına
    -0.06
    (field
    -0.06
    VD
    -0.05
    groupName
    -0.05
    POSITIVE LOGITS
    .Resource
    0.07
     unaffected
    0.07
    ection
    0.07
    ucceed
    0.07
    greSQL
    0.07
     Edit
    0.06
    Edit
    0.06
     mayoría
    0.06
     crem
    0.06
    tober
    0.06
    Activations Density 0.012%

    No Known Activations