© 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. 7857
    Prev
    Next
    INDEX
    Explanations

    "ine" (drug suffix)

    np_max-act · gemini-2.0-flash

    The neuron is especially sensitive to drug names ending in “-dipine,” i.e. dihydropyridine calcium‐channel blockers.

    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
    InChildren
    -0.07
     Hwy
    -0.06
    Format
    -0.06
     {.
    -0.06
     Penn
    -0.06
    (pd
    -0.06
     geom
    -0.06
     tumors
    -0.06
    '{
    -0.06
     '{
    -0.06
    POSITIVE LOGITS
    اظ
    0.08
     ανά
    0.07
    реп
    0.07
     rif
    0.07
    sizlik
    0.07
    osate
    0.07
     σχέ
    0.07
    oline
    0.07
     Νο
    0.07
    бина
    0.07
    Activations Density 0.079%

    No Known Activations