© 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. 6323
    Prev
    Next
    INDEX
    Explanations

    font

    np_max-act · gemini-2.0-flash

    The neuron fires on CSS tokens related to FontAwesome—e.g. class names, mixin variables, and pseudo-elements prefixed “fa” or “FontAwesome.”

    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
     CE
    -0.07
     NR
    -0.07
     RD
    -0.06
    щается
    -0.06
    �
    -0.06
    _layer
    -0.06
     mosque
    -0.06
     非
    -0.06
    _non
    -0.06
     fadeIn
    -0.06
    POSITIVE LOGITS
     Drawing
    0.07
     Assault
    0.07
     reiterated
    0.07
     Division
    0.06
    _customize
    0.06
    (())↵
    0.06
     amended
    0.06
    <()>
    0.06
     andre
    0.06
    -notes
    0.06
    Activations Density 0.011%

    No Known Activations