© 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
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 13589
    Prev
    Next
    INDEX
    Explanations

    Written material

    np_max-act · gemini-2.0-flash

    This neuron detects instruction-like phrases that reference or point to the “following passage,” such as “based on the following passage.”

    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
     мист
    -0.07
     nightmare
    -0.06
    aad
    -0.06
    AG
    -0.06
    american
    -0.06
    nx
    -0.06
    QR
    -0.06
    _Final
    -0.06
     elephant
    -0.06
     террит
    -0.06
    POSITIVE LOGITS
    رة
    0.07
     ",
    ↵
    0.07
     дів
    0.07
    กล
    0.06
    >";
    ↵
    0.06
    sorry
    0.06
    Opens
    0.06
    APolynomial
    0.06
    ruby
    0.06
     newline
    0.06
    Activations Density 0.024%

    No Known Activations