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. 40428
    Prev
    Next
    INDEX
    Explanations

    the, with

    np_max-act · gemini-2.0-flash

    company information focused on chemical products and services.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron fires on boilerplate introductory phrases that begin with “with” (e.g. “with the aim/goal of,” “with a focus on”), marking those “with … ” clauses.

    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
    Fortunately
    -0.07
    众
    -0.06
    ові
    -0.06
    Fs
    -0.06
    noticed
    -0.06
    rát
    -0.06
    "nil
    -0.06
    acles
    -0.06
     변화
    -0.06
    -Version
    -0.06
    POSITIVE LOGITS
     라이
    0.06
     холод
    0.06
     листоп
    0.06
     merchant
    0.06
    YD
    0.06
     باد
    0.06
    (train
    0.06
     والم
    0.06
    ยาน
    0.06
     flights
    0.06
    Activations Density 0.011%

    No Known Activations