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. 86012
    Prev
    Next
    INDEX
    Explanations

    dating/relationships

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of present-perfect verb constructions (e.g. "has … remarried").

    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
    .ComponentResourceManager
    -0.06
     antenn
    -0.06
     уточ
    -0.06
     waiting
    -0.06
     '/');↵
    -0.06
     kode
    -0.06
     技
    -0.06
    Latch
    -0.06
     опас
    -0.06
    arms
    -0.06
    POSITIVE LOGITS
    <State
    0.07
     ENGINE
    0.07
    .removeListener
    0.06
     TREE
    0.06
    ameleon
    0.06
     Rack
    0.06
     cabel
    0.06
    rip
    0.06
     AppBar
    0.06
    рі
    0.06
    Activations Density 0.037%

    No Known Activations