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. 88870
    Prev
    Next
    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric literals (e.g. line numbers, error codes, version or size values) in code and compiler messages.

    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
     sert
    -0.07
    Meg
    -0.07
    色的
    -0.07
    vanized
    -0.06
    (te
    -0.06
    Tipo
    -0.06
    imization
    -0.06
     pumping
    -0.06
    кот
    -0.06
     件
    -0.06
    POSITIVE LOGITS
    resents
    0.07
     torque
    0.07
     rand
    0.06
     Inspir
    0.06
    	conn
    0.06
    plain
    0.06
     tako
    0.06
    จ
    0.06
    illum
    0.06
    amous
    0.06
    Activations Density 0.002%

    No Known Activations