© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 125032
    Prev
    Next
    INDEX
    Explanations

    tongue

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens naming or describing mouth-and-tongue sexual contact (e.g. “tongue,” “licking,” “kissing”).

    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
    梦
    -0.07
     кле
    -0.07
     IF
    -0.07
     LOT
    -0.06
     bins
    -0.06
     ]}↵
    -0.06
    GetComponent
    -0.06
     approximately
    -0.06
    58
    -0.06
     capacitor
    -0.06
    POSITIVE LOGITS
     tongue
    0.13
     tongues
    0.09
     Tong
    0.07
     tyranny
    0.07
     compuls
    0.07
     touchscreen
    0.07
    .Time
    0.07
    _eng
    0.07
     Turkey
    0.06
     Commentary
    0.06
    Activations Density 0.004%

    No Known Activations