© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 99705
    Prev
    Next
    INDEX
    Explanations

    emergencies, warnings, and calls for help

    np_max-act · gemini-2.0-flash

    whether the text is expressing a safety concern or giving urgent safety-related advice.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0

    safety warnings and guidance that advise leaving the area and contacting professionals/authorities in response to a dangerous situation.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Feedback
    -0.07
     Sue
    -0.07
     pull
    -0.06
     organisations
    -0.06
     рівня
    -0.06
     evolves
    -0.06
     racial
    -0.06
     participate
    -0.06
    proved
    -0.06
     highlights
    -0.06
    POSITIVE LOGITS
     chiar
    0.07
    prise
    0.07
    ẩu
    0.06
     уяв
    0.06
    acie
    0.06
    chan
    0.06
    	mutex
    0.06
    	↵	↵	↵	↵
    0.06
     İŞ
    0.06
     суще
    0.06
    Activations Density 0.020%

    No Known Activations