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. 40405
    Prev
    Next
    INDEX
    Explanations

    punctuation and "and"

    np_max-act · gemini-2.0-flash

    The neuron detects floating‐point numeric values (decimal numbers) in the text.

    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
    /cpu
    -0.07
    .jface
    -0.07
    xc
    -0.07
     service
    -0.07
    放送
    -0.07
     bureau
    -0.07
    ople
    -0.07
    #!
    -0.06
    >--}}↵
    -0.06
    017
    -0.06
    POSITIVE LOGITS
     dafür
    0.07
     forwarded
    0.06
     GridView
    0.06
     Float
    0.06
     PARA
    0.06
     HLS
    0.06
     Ravens
    0.06
     поступ
    0.06
     valeurs
    0.06
     पड़
    0.05
    Activations Density 0.148%

    No Known Activations