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. 86044
    Prev
    Next
    INDEX
    Explanations

    K

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on mentions of extracellular signal–regulated kinase (ERK) and related MAP kinase pathway components.

    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
    ้อน
    -0.08
     tbody
    -0.06
    adia
    -0.06
    AA
    -0.06
    hodob
    -0.06
    πα
    -0.06
    OnError
    -0.06
     Pandora
    -0.06
     Mob
    -0.06
     insol
    -0.06
    POSITIVE LOGITS
    _dbg
    0.07
     Sporting
    0.07
    ,[
    0.07
     <>
    0.06
     supply
    0.06
    Swagger
    0.06
     Spanish
    0.06
    gre
    0.06
     tool
    0.06
     Mage
    0.06
    Activations Density 0.003%

    No Known Activations