Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 85877
    Prev
    Next
    INDEX
    Explanations

    carrot

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions or descriptions of root vegetables—specifically on words like “root,” “tuberous,” or “bulbous” that identify a vegetable as a root.

    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
    tgl
    -0.06
    나는
    -0.06
    lda
    -0.06
     getCurrent
    -0.06
    ियत
    -0.06
     тиж
    -0.06
    ZeroWidthSpace
    -0.06
    .Tween
    -0.06
    agement
    -0.06
    inema
    -0.06
    POSITIVE LOGITS
     carrots
    0.10
    rv
    0.08
     carrot
    0.08
    cv
    0.07
     �
    0.07
    -r
    0.07
     straight
    0.07
    奈
    0.07
    trained
    0.06
     heroin
    0.06
    Activations Density 0.003%

    No Known Activations