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. 40475
    Prev
    Next
    INDEX
    Explanations

    Avoidance

    np_max-act · gemini-2.0-flash

    The neuron fires on imperative or cautionary verbs and gerunds in safety‐oriented instructions (e.g. “avoid,” “using,” “prolonged,” “exposure,” 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
     Snapdragon
    -0.07
    Cliente
    -0.07
    Matcher
    -0.07
    Reward
    -0.06
     sứ
    -0.06
    (objects
    -0.06
     plagiarism
    -0.06
     Oy
    -0.06
    interop
    -0.06
    参加
    -0.06
    POSITIVE LOGITS
    uario
    0.07
    ูล
    0.07
    KD
    0.07
    itchen
    0.06
    rots
    0.06
     taken
    0.06
     треб
    0.06
    rients
    0.06
    (resp
    0.06
    orig
    0.06
    Activations Density 0.061%

    No Known Activations