© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 88949
    Prev
    Next
    INDEX
    Explanations

    applying for something

    np_max-act · gemini-2.0-flash

    information related to U.S. immigration and visa processes.

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

    The neuron specifically activates on forms of the word “apply” (including “application,” “applying,” etc.), i.e. tokens related to applying.

    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
    ُس
    -0.07
     Coding
    -0.07
     donner
    -0.06
     Tort
    -0.06
     nudity
    -0.06
    ials
    -0.06
     Labor
    -0.06
     buses
    -0.06
    _Session
    -0.06
    _nb
    -0.06
    POSITIVE LOGITS
     applying
    0.09
     apply
    0.08
     applicant
    0.07
     application
    0.07
     Apply
    0.07
    Entering
    0.07
     lw
    0.07
    paralleled
    0.07
    administration
    0.07
    .newBuilder
    0.07
    Activations Density 0.020%

    No Known Activations