© 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. 126072
    Prev
    Next
    INDEX
    Explanations

    sports players

    np_max-act · gemini-2.0-flash

    The neuron mainly fires on structural or formatting markers—section/header tokens and list‐item indicators (e.g. numbered markers, special header IDs).

    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
     pepper
    -0.07
    .square
    -0.07
    428
    -0.06
    _FLOAT
    -0.06
    :t
    -0.06
    	Set
    -0.06
     국제
    -0.06
     skin
    -0.06
     deutschen
    -0.06
    .au
    -0.06
    POSITIVE LOGITS
     marca
    0.07
    Sha
    0.06
     čty
    0.06
    []>(
    0.06
     उन
    0.06
     diagrams
    0.06
    Grammar
    0.06
    ूछ
    0.06
    neh
    0.06
    	RTLR
    0.06
    Activations Density 0.032%

    No Known Activations