© 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. 68863
    Prev
    Next
    INDEX
    Explanations

    outfield

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of a baseball outfielder position (e.g., “outfield,” “left field,” “right field”).

    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
     polluted
    -0.07
    finance
    -0.06
     기준
    -0.06
     Friday
    -0.06
     shipped
    -0.06
    electronics
    -0.06
    ávis
    -0.06
     cancellation
    -0.06
     Tuesday
    -0.06
    cház
    -0.06
    POSITIVE LOGITS
    	GL
    0.07
     VF
    0.07
    Orden
    0.07
     nymph
    0.07
     pł
    0.07
    :L
    0.06
    (([
    0.06
     holog
    0.06
     TFT
    0.06
    یتی
    0.06
    Activations Density 0.001%

    No Known Activations