Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 84638
    Prev
    Next
    INDEX
    Explanations

    dt

    np_max-act · gemini-2.0-flash

    The neuron detects occurrences of the “org.eclipse.jdt.core” package identifier (i.e. the JDT Core namespace).

    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
    ροι
    -0.06
     gradients
    -0.06
    цій
    -0.06
    memcpy
    -0.06
    emand
    -0.06
    ckeditor
    -0.06
     нат
    -0.06
     gỗ
    -0.06
    شور
    -0.06
    ektor
    -0.06
    POSITIVE LOGITS
     підстав
    0.07
     relocated
    0.06
    updated
    0.06
    -direct
    0.06
    .fail
    0.06
     extends
    0.06
    ='%
    0.06
    .created
    0.06
    ْد
    0.06
    .slice
    0.06
    Activations Density 0.000%

    No Known Activations