© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 43126
    Prev
    Next
    INDEX
    Explanations

    unknown or little-known

    np_max-act · gemini-2.0-flash

    This neuron activates on words and short phrases that signal a subject’s familiarity or obscurity—e.g. “unknown,” “household name,” “never heard,” “you know who she is.”

    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
     steal
    -0.07
     سفر
    -0.07
     각
    -0.07
     glaciers
    -0.07
     Lawn
    -0.06
    -rating
    -0.06
     down
    -0.06
    かわ
    -0.06
    された
    -0.06
     wand
    -0.06
    POSITIVE LOGITS
    _GREEN
    0.06
    .sel
    0.06
     غذ
    0.06
    Arrays
    0.06
    =max
    0.06
    ableOpacity
    0.06
    _NEED
    0.06
    _definition
    0.06
    omore
    0.06
     모르
    0.06
    Activations Density 0.036%

    No Known Activations