© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 30075
    Prev
    Next
    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    This neuron detects tokens that specify response‐format instructions, like “table,” “sentiments,” and “Do.”

    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
     моя
    -0.07
    า
    -0.07
     edilmiş
    -0.07
    .readString
    -0.06
    Radians
    -0.06
    .toastr
    -0.06
    ilmiş
    -0.06
     orient
    -0.06
    .car
    -0.06
     роб
    -0.06
    POSITIVE LOGITS
    641
    0.07
     goats
    0.06
    část
    0.06
    829
    0.06
    nova
    0.06
    nit
    0.06
    Ne
    0.06
    ЎыџN
    0.06
     vital
    0.06
     PAGE
    0.06
    Activations Density 0.005%

    No Known Activations