© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 43297
    Prev
    Next
    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric literals—especially pixel dimension values (e.g. the numbers before “px”) in CSS.

    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
    .tmp
    -0.07
     účin
    -0.06
    所属
    -0.06
     klíč
    -0.06
     sizes
    -0.06
    _rad
    -0.06
    Sing
    -0.06
     suis
    -0.06
     пуст
    -0.06
    ']."'
    -0.06
    POSITIVE LOGITS
    .equalsIgnoreCase
    0.08
    quelle
    0.07
     그냥
    0.07
    getName
    0.07
     �
    0.06
     Mozilla
    0.06
    idebar
    0.06
    elts
    0.06
    <body
    0.06
    анка
    0.06
    Activations Density 0.012%

    No Known Activations