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. 86242
    Prev
    Next
    INDEX
    Explanations

    medical illnesses/disease

    np_max-act · gemini-2.0-flash

    The neuron responds to words that signal a relationship or causal/underlying link between concepts (e.g. “relationship,” “temporal,” “underlying”).

    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
    ительность
    -0.08
    nout
    -0.06
     SIZE
    -0.06
    enie
    -0.06
     Qualität
    -0.06
    ملة
    -0.06
     Cunningham
    -0.06
    Miller
    -0.06
    OID
    -0.06
    kf
    -0.06
    POSITIVE LOGITS
     reflected
    0.07
    근
    0.07
     精
    0.07
    <'
    0.06
     Госп
    0.06
    丰
    0.06
     rust
    0.06
     blush
    0.06
     phụ
    0.06
    }")↵
    0.06
    Activations Density 0.153%

    No Known Activations