© 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. 59011
    Prev
    Next
    INDEX
    Explanations

    ill

    np_max-act · gemini-2.0-flash

    The neuron fires on content‐bearing tokens—especially nouns and verbs naming organizations, groups, roles, or technical concepts—i.e. the key semantic “topic” words in an informational passage.

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

    words related to the structure and characteristics of organizations, particularly with respect to cults and their operations.

    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
     minute
    -0.06
    ailed
    -0.06
    ivers
    -0.06
    ater
    -0.06
     twin
    -0.06
     abandoning
    -0.06
     principle
    -0.06
    ATER
    -0.06
    afür
    -0.06
     COMPUTER
    -0.06
    POSITIVE LOGITS
     stabilized
    0.07
    .snap
    0.07
    inou
    0.07
    agr
    0.06
     MPG
    0.06
     Chern
    0.06
    ibilidade
    0.06
    /includes
    0.06
     foss
    0.06
     sortOrder
    0.06
    Activations Density 0.400%

    No Known Activations