© 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. 126735
    Prev
    Next
    INDEX
    Explanations

    descriptive storytelling

    np_max-act · gemini-2.0-flash

    The neuron responds to tokens that describe characters’ internal physical sensations or emotional reactions (like “jerking,” “gasped,” “muffled,” “laughter,” etc.).

    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
     Verd
    -0.07
     fd
    -0.07
     concession
    -0.07
     counterfeit
    -0.07
    apatkan
    -0.07
    ilan
    -0.06
     distributors
    -0.06
    Occup
    -0.06
     ApiController
    -0.06
     veto
    -0.06
    POSITIVE LOGITS
    prevent
    0.06
    (DateTime
    0.06
    although
    0.06
     collaborating
    0.06
     torch
    0.06
     mattresses
    0.06
    .Information
    0.06
    การท
    0.06
     г
    0.06
     обличчя
    0.06
    Activations Density 0.019%

    No Known Activations