© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 101422
    Prev
    Next
    INDEX
    Explanations

    @

    np_max-act · gemini-2.0-flash

    The neuron specializes in spotting shell‐style command prompts, i.e. tokens of the form “username@hostname” you see in terminal sessions.

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

    commands and issues related to scripting or programming in a Linux environment.

    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
    'en
    -0.07
    Năm
    -0.07
     Cabinets
    -0.06
    ahi
    -0.06
     blk
    -0.06
    Č
    -0.06
    ้ด
    -0.06
    คาส
    -0.06
    xf
    -0.06
    ')}}"
    -0.06
    POSITIVE LOGITS
    inois
    0.07
    pq
    0.07
    crets
    0.07
     counterparts
    0.07
     polov
    0.06
    _sqrt
    0.06
    nonatomic
    0.06
    +B
    0.06
     rebut
    0.06
    acted
    0.06
    Activations Density 0.002%

    No Known Activations