Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 109605
    Prev
    Next
    INDEX
    Explanations

    fragmented web text

    np_max-act · gemini-2.0-flash

    This neuron activates on decimal number tokens (fractional score values) in the table.

    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
     onBind
    -0.06
     embr
    -0.06
     skiing
    -0.06
    _walk
    -0.06
    。今
    -0.06
    Relations
    -0.06
    شمالی
    -0.06
     poil
    -0.06
    methodPointerType
    -0.05
     effected
    -0.05
    POSITIVE LOGITS
    Book
    0.07
     buys
    0.07
    喜
    0.07
    ılmaz
    0.06
     did
    0.06
     capacidad
    0.06
     }]);↵
    0.06
    kyně
    0.06
    ющими
    0.06
     knocking
    0.06
    Activations Density 0.002%

    No Known Activations