Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 130652
    Prev
    Next
    INDEX
    Explanations

    lips

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of “lips” (i.e. the word “lips” and its immediate descriptive context).

    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
    orough
    -0.07
     dirty
    -0.06
    Seen
    -0.06
     Fey
    -0.06
    176
    -0.06
     GtkWidget
    -0.06
     Roh
    -0.06
     arithmetic
    -0.06
     Fuk
    -0.06
     �
    -0.06
    POSITIVE LOGITS
     lips
    0.15
     lip
    0.10
     Lips
    0.09
    ilip
    0.09
     Lip
    0.09
     lipstick
    0.08
    IP
    0.08
    ip
    0.08
    IPPING
    0.08
    args
    0.08
    Activations Density 0.005%

    No Known Activations