© 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. 11-RESID-POST-AA
    11. 6810
    Prev
    Next
    INDEX
    Explanations

    drinking

    np_max-act · gemini-2.0-flash

    The neuron activates on any form of the word “drink” (e.g. drink, drinking, drinks).

    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
     Owen
    -0.09
     Xavier
    -0.07
     Mesa
    -0.07
     expo
    -0.07
     Pad
    -0.07
    grid
    -0.07
     Ted
    -0.07
    aaa
    -0.07
     axis
    -0.07
     Percy
    -0.07
    POSITIVE LOGITS
     drink
    0.11
     drinking
    0.11
     Drinking
    0.09
     Drink
    0.09
    drink
    0.08
     drank
    0.08
     Drinks
    0.08
    饮
    0.08
    Drink
    0.08
    りの
    0.08
    Activations Density 0.010%

    No Known Activations