© 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. 55452
    Prev
    Next
    INDEX
    Explanations

    tri

    np_max-act · gemini-2.0-flash

    The neuron fires on short code- or filename-style chunks (e.g. fragments of library routine names like “dsptri” and file extensions such as “.zip” or “.txt”).

    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
    ITTE
    -0.07
    already
    -0.06
    that
    -0.06
    overs
    -0.06
    伙
    -0.06
    írk
    -0.06
    (route
    -0.06
    $
    ↵
    -0.06
    with
    -0.06
    ố
    -0.06
    POSITIVE LOGITS
     xm
    0.07
     treeNode
    0.07
     goggles
    0.07
     pochop
    0.07
     CSL
    0.06
     Boat
    0.06
     Ск
    0.06
     Coy
    0.06
    notated
    0.06
     इसस
    0.06
    Activations Density 0.171%

    No Known Activations