© 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 EvalsExportsinterp-engineNEWAPI 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. 102527
    Prev
    Next
    INDEX
    Explanations

    die

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on words denoting death, especially the verb “die” and its “-ing” form (“dying”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    expressions of despair and existential dread.

    oai_token-act-pair · gpt-4o-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
     grab
    -0.07
     compute
    -0.06
     克
    -0.06
     Fetch
    -0.06
     shorts
    -0.06
     explored
    -0.06
    běhu
    -0.06
     short
    -0.06
     neatly
    -0.06
    =req
    -0.06
    POSITIVE LOGITS
     died
    0.13
     dying
    0.12
     dies
    0.10
     die
    0.10
     Died
    0.09
     Die
    0.09
    死
    0.08
     Dies
    0.08
     born
    0.08
    ダイ
    0.07
    Activations Density 0.010%

    No Known Activations