© 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. 44861
    Prev
    Next
    INDEX
    Explanations

    Negativity

    np_max-act · gemini-2.0-flash

    This neuron detects descriptions of someone approaching or interacting with “you,” especially when they seem friendly at first but then ignore or disappear.

    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
    observe
    -0.07
     Horizon
    -0.07
     Leap
    -0.07
    ako
    -0.07
     defamation
    -0.07
    	exports
    -0.07
    -0.06
    bsites
    -0.06
     walked
    -0.06
     अब
    -0.06
    POSITIVE LOGITS
    .Il
    0.06
    摆
    0.06
     Minister
    0.06
    resse
    0.06
     أغ
    0.06
    aphael
    0.06
    razil
    0.06
    จำก
    0.06
    울
    0.06
    igidBody
    0.05
    Activations Density 0.024%

    No Known Activations