© 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. 38898
    Prev
    Next
    INDEX
    Explanations

    kindness

    np_max-act · gemini-2.0-flash

    The neuron detects words and phrases expressing a character’s emotional soft side or vulnerability (e.g., “soft spot,” “loyal,” “vulnerability”).

    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
     ram
    -0.07
    君
    -0.06
    २०
    -0.06
    黒
    -0.06
     works
    -0.06
    идент
    -0.06
     tbody
    -0.06
     functionName
    -0.06
    िज
    -0.06
     Hawks
    -0.06
    POSITIVE LOGITS
     Nursery
    0.07
    _NAMESPACE
    0.07
     Numerous
    0.07
     suggested
    0.07
    Mutable
    0.07
     कभ
    0.07
     multiplier
    0.06
    (EXPR
    0.06
    stood
    0.06
     thiện
    0.06
    Activations Density 0.026%

    No Known Activations