© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 9114
    Prev
    Next
    INDEX
    Explanations

    disclaimers and information

    np_max-act · gemini-2.0-flash

    The neuron detects hedging or citation language—words that flag approximate figures or attribute data to sources (e.g. “based,” “sourced,” “reliable,” “approximate”).

    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
     ECM
    -0.07
    _sim
    -0.06
     maze
    -0.06
    ULAR
    -0.06
     load
    -0.06
     trustees
    -0.06
    endon
    -0.05
     soy
    -0.05
     ROUND
    -0.05
     Sonuç
    -0.05
    POSITIVE LOGITS
    ata
    0.07
    	mutex
    0.06
     Leslie
    0.06
    &view
    0.06
    .OS
    0.06
     hö
    0.06
    спіль
    0.06
    faf
    0.06
     Gä
    0.06
    향
    0.06
    Activations Density 0.026%

    No Known Activations