© 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
    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. 73712
    Prev
    Next
    INDEX
    Explanations

    negation

    np_max-act · gemini-2.0-flash

    The neuron specializes in spotting statements about not needing or requiring something (e.g. “does not require any…,” “requires no…”).

    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
    liğini
    -0.07
     eens
    -0.07
     Jacques
    -0.07
    adecimal
    -0.06
     mqtt
    -0.06
     Ihrem
    -0.06
    oloji
    -0.06
    tsx
    -0.06
     più
    -0.06
    addons
    -0.06
    POSITIVE LOGITS
    ANGE
    0.07
     civilization
    0.07
     GIR
    0.06
     insulation
    0.06
    عات
    0.06
     digestion
    0.06
    igraphy
    0.06
     }]↵
    0.06
     enrichment
    0.06
     😉↵↵
    0.06
    Activations Density 0.038%

    No Known Activations