© 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. 6500
    Prev
    Next
    INDEX
    Explanations

    arial

    np_max-act · gemini-2.0-flash

    The neuron activates on the word “adversarial,” particularly in the phrase “generative adversarial.”

    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
     whitelist
    -0.07
    くな
    -0.07
     ><
    -0.06
     보내
    -0.06
     الموقع
    -0.06
     poetic
    -0.06
     plainly
    -0.06
    れない
    -0.06
     SelectList
    -0.06
     glaciers
    -0.06
    POSITIVE LOGITS
    анная
    0.07
     Bro
    0.07
     indispens
    0.06
     An
    0.06
    arl
    0.06
     Sm
    0.06
     Bord
    0.06
     CHR
    0.06
    _control
    0.06
     Gabri
    0.06
    Activations Density 0.000%

    No Known Activations