© 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. 126614
    Prev
    Next
    INDEX
    Explanations

    erotic descriptions

    np_max-act · gemini-2.0-flash

    The neuron responds to the placeholder tokens used for character identifiers (e.g. “NAME_1,” “NAME_3,” the trailing digits, etc.).

    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
    (home
    -0.07
     نص
    -0.06
     muscles
    -0.06
    .finish
    -0.06
    .when
    -0.06
    soft
    -0.06
    _Instance
    -0.06
    _sms
    -0.06
    asm
    -0.06
    	seq
    -0.06
    POSITIVE LOGITS
     ulus
    0.06
     Campo
    0.06
     บ
    0.06
    -folder
    0.06
     القد
    0.06
     відбува
    0.06
    への
    0.06
    (HttpContext
    0.06
    olib
    0.06
     الإن
    0.06
    Activations Density 0.024%

    No Known Activations