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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 88278
    Prev
    Next
    INDEX
    Explanations

    coding algorithms

    np_max-act · gemini-2.0-flash

    data structures used in the solution code.

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

    The neuron fires on tokens that are part of algorithm‐problem identifiers or function names (e.g. “Duplicate,” “Consecutive,” “Matrix,” “Combinations,” “Zero,” etc.).

    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
     리스트
    -0.07
     Copyright
    -0.06
    JSONArray
    -0.06
     визнач
    -0.06
     حتی
    -0.06
     删除
    -0.06
    _CATEGORY
    -0.06
    ुर
    -0.06
    WhiteSpace
    -0.06
     Dialogue
    -0.06
    POSITIVE LOGITS
    .Progress
    0.07
    mah
    0.06
    actor
    0.06
    	state
    0.06
     interoper
    0.06
    .Symbol
    0.06
    Mari
    0.06
    .plan
    0.05
    inker
    0.05
    .HtmlControls
    0.05
    Activations Density 0.010%

    No Known Activations