© 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
    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. 73995
    Prev
    Next
    INDEX
    Explanations

    Reviews

    np_max-act · gemini-2.0-flash

    The neuron is highly sensitive to content words that are action verbs (e.g. verbs describing events, developments, or narrative progression).

    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
    (TIM
    -0.06
    .required
    -0.06
     sluts
    -0.06
    dense
    -0.06
    785
    -0.06
    _Api
    -0.06
     airstrikes
    -0.06
    Shape
    -0.06
     thick
    -0.06
     nacional
    -0.06
    POSITIVE LOGITS
    主题
    0.06
    OLUTE
    0.06
    运行
    0.06
    ushman
    0.06
    �프
    0.06
    'utilisation
    0.06
    (address
    0.06
    Jun
    0.06
     Orlando
    0.06
     "__
    0.06
    Activations Density 0.059%

    No Known Activations