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. 40538
    Prev
    Next
    INDEX
    Explanations

    HTML/CSS code

    np_max-act · gemini-2.0-flash

    This neuron responds to decimal number tokens (i.e. floating‐point values) in the CSS.

    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
    _____
    -0.07
     \@
    -0.06
    	prev
    -0.06
     guns
    -0.06
     ren
    -0.06
    tam
    -0.06
    �
    -0.06
     pager
    -0.06
    /item
    -0.06
    xfb
    -0.06
    POSITIVE LOGITS
    Пол
    0.08
     законом
    0.07
    InstantiationException
    0.07
    conn
    0.07
     coveted
    0.07
     Callback
    0.06
     AL
    0.06
    Book
    0.06
    21
    0.06
     Sending
    0.06
    Activations Density 0.003%

    No Known Activations