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. 81385
    Prev
    Next
    INDEX
    Explanations

    column

    np_max-act · gemini-2.0-flash

    This neuron never fires on any meaningful token sequence—it effectively detects “nothing” (i.e. only whitespace or uninformative padding) and does not activate on real content.

    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
    ANCELED
    -0.07
    ellt
    -0.06
    _documento
    -0.06
    top
    -0.06
    Icon
    -0.06
    _STARTED
    -0.06
    851
    -0.06
    slack
    -0.06
     compliment
    -0.06
    ../../../
    -0.06
    POSITIVE LOGITS
    :auto
    0.07
    場
    0.07
    conduct
    0.06
    (beta
    0.06
     美
    0.06
    thane
    0.06
     sire
    0.06
    :text
    0.06
     RU
    0.06
     Dough
    0.06
    Activations Density 0.003%

    No Known Activations