© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 92150
    Prev
    Next
    INDEX
    Explanations

    stalls

    np_max-act · gemini-2.0-flash

    The neuron fires on formal nouns and terms naming institutions, bureaucratic bodies, or official actions (e.g. “committee,” “office,” “operations,” “minister,” “articles”).

    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
    sig
    -0.06
     atoi
    -0.06
    izzard
    -0.06
    γκο
    -0.06
    _subset
    -0.06
    .cells
    -0.06
    ye
    -0.06
    (t
    -0.06
     during
    -0.06
    企
    -0.06
    POSITIVE LOGITS
    "]=
    0.08
    =<
    0.07
    :normal
    0.07
    _MODAL
    0.07
     clearInterval
    0.06
    -medium
    0.06
     Morton
    0.06
     Contractor
    0.06
    bands
    0.06
    PUT
    0.06
    Activations Density 0.150%

    No Known Activations