© 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. 41567
    Prev
    Next
    INDEX
    Explanations

    disrespect

    np_max-act · gemini-2.0-flash

    The neuron flags terms that explicitly label content as disrespectful, derogatory, or offensive.

    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
     bình
    -0.07
     rehabilitation
    -0.07
     consolation
    -0.06
     xi
    -0.06
    .Custom
    -0.06
     Statistics
    -0.06
     wildcard
    -0.06
     nhu
    -0.06
    .Fail
    -0.06
    _Form
    -0.06
    POSITIVE LOGITS
     disrespectful
    0.08
     
    0.07
    ẹ
    0.07
     viết
    0.06
     Datum
    0.06
     hateful
    0.06
    .depth
    0.06
    墓
    0.06
    ональ
    0.06
     cler
    0.06
    Activations Density 0.009%

    No Known Activations