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

    Neuronpedia

    Jacobian LensNEW
    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. 39015
    Prev
    Next
    INDEX
    Explanations

    relationship

    np_max-act · gemini-2.0-flash

    This neuron detects phrases that state or introduce a “relationship between X and Y,” i.e. constructions explicitly framing a connection between two scientific variables or concepts.

    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
    08
    -0.09
    04
    -0.08
    048
    -0.08
    07
    -0.07
    10
    -0.07
    18
    -0.07
    05
    -0.07
    03
    -0.07
    ->_
    -0.07
    23
    -0.07
    POSITIVE LOGITS
     relationship
    0.13
     relationships
    0.12
     Relationship
    0.11
     Relationships
    0.10
    relationship
    0.10
    �체
    0.09
     رابطه
    0.08
    Relationship
    0.08
    relationships
    0.08
    returnValue
    0.08
    Activations Density 0.019%

    No Known Activations