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. 129070
    Prev
    Next
    INDEX
    Explanations

    cell cultures, lab experiments

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of cell‐culture assay fluids—words like “culture,” “medium,” “media,” and “supernatant.”

    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
     turns
    -0.08
    ゾ
    -0.07
     rectangles
    -0.07
    Psi
    -0.07
    冊
    -0.07
    Coder
    -0.07
     ध
    -0.06
     System
    -0.06
     злоч
    -0.06
     Systems
    -0.06
    POSITIVE LOGITS
     front
    0.07
     dưỡng
    0.07
     hepat
    0.06
    	wg
    0.06
     solitude
    0.06
     massage
    0.06
    -themed
    0.06
    _visited
    0.06
     head
    0.06
     کیف
    0.06
    Activations Density 0.005%

    No Known Activations