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. 84476
    Prev
    Next
    INDEX
    Explanations

    FILE

    np_max-act · gemini-2.0-flash

    This neuron is triggered by the uppercase token “FILE,” i.e. it spots the “FILE” marker at the top of court‐document headers.

    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
     Ди
    -0.07
     attracts
    -0.07
     nye
    -0.07
     segu
    -0.07
    _POLL
    -0.06
     относ
    -0.06
     RECEIVE
    -0.06
     pleasure
    -0.06
     control
    -0.06
     mereka
    -0.06
    POSITIVE LOGITS
     feasible
    0.06
     Marty
    0.06
    issue
    0.06
     生命周期函数
    0.06
     â
    0.06
     Instantiate
    0.06
     Called
    0.06
    방법
    0.06
    chner
    0.06
    irt
    0.06
    Activations Density 0.000%

    No Known Activations