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. 130605
    Prev
    Next
    INDEX
    Explanations

    poetry/literature

    np_max-act · gemini-2.0-flash

    This neuron consistently activates on the very first word of each line or stanza, marking line-initial tokens.

    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
     insulated
    -0.07
    ordon
    -0.07
    ifecycle
    -0.07
     intact
    -0.06
     wired
    -0.06
     potent
    -0.06
     fac
    -0.06
    clinic
    -0.06
     pharmaceutical
    -0.06
    達
    -0.06
    POSITIVE LOGITS
    "]');↵
    0.06
     Rodrigo
    0.06
     mContext
    0.06
     مط
    0.06
     tossing
    0.06
    <Group
    0.06
    _HIT
    0.06
     Item
    0.06
     Area
    0.06
    Rates
    0.06
    Activations Density 0.016%

    No Known Activations