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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 91635
    Prev
    Next
    INDEX
    Explanations

    providing information

    np_max-act · gemini-2.0-flash

    This neuron detects numbered list markers (e.g. “1.”, “2.”, “3.”) that denote steps or items in a procedure.

    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
     받아
    -0.07
    Cl
    -0.06
     informat
    -0.06
     Thief
    -0.06
    nehmen
    -0.06
    idian
    -0.06
    estimate
    -0.06
    나요
    -0.06
     holog
    -0.06
     Memories
    -0.06
    POSITIVE LOGITS
    ويك
    0.07
     três
    0.06
     MATLAB
    0.06
     万
    0.06
     artikel
    0.06
    wrapper
    0.06
    	do
    0.06
     afect
    0.06
     Spir
    0.06
    698
    0.06
    Activations Density 0.022%

    No Known Activations