© 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. 934
    Prev
    Next
    INDEX
    Explanations

    funny

    np_max-act · gemini-2.0-flash

    The neuron detects humor-related descriptors—words like “funny,” “hum,” “humorous,” or similar markers of comedic tone.

    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
    уков
    -0.07
     umbrella
    -0.07
     Alone
    -0.06
    .accept
    -0.06
     Scientist
    -0.06
     squirt
    -0.06
     exploitation
    -0.06
     worthwhile
    -0.06
     prenatal
    -0.06
    沿
    -0.06
    POSITIVE LOGITS
     hilarious
    0.08
     exposition
    0.08
     newspaper
    0.07
     Ep
    0.07
     verbs
    0.07
    	Add
    0.06
     CircularProgressIndicator
    0.06
     ense
    0.06
    \HttpFoundation
    0.06
    NDER
    0.06
    Activations Density 0.021%

    No Known Activations