© 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. 9061
    Prev
    Next
    INDEX
    Explanations

    General text snippets

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens representing decimal or floating‐point values (e.g. “0.22509765625,” “1.66796875,” etc.).

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

    tokens that begin sentences or paragraphs (sentence/section-initial words, including headings).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
     hamburg
    -0.08
    evenodd
    -0.07
     builder
    -0.07
     celular
    -0.07
    .depart
    -0.07
     sổ
    -0.06
    .parentElement
    -0.06
    Cd
    -0.06
    -left
    -0.06
     everlasting
    -0.06
    POSITIVE LOGITS
     VIDEO
    0.07
     estão
    0.06
     speculated
    0.06
    ERVE
    0.06
     pagar
    0.06
     Double
    0.06
     المتحدة
    0.06
    Interface
    0.06
     doubt
    0.06
    违
    0.06
    Activations Density 0.293%

    No Known Activations