© 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. 126349
    Prev
    Next
    INDEX
    Explanations

    ic

    np_max-act · gemini-2.0-flash

    The neuron fires on the “ic” in category labels like “Olympic,” i.e. it detects the “-ic” substring in words such as “Olympic.”

    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
     assertThat
    -0.07
     limited
    -0.07
    Ó
    -0.06
    чний
    -0.06
     Sum
    -0.06
    FetchRequest
    -0.06
    edores
    -0.06
     داشت
    -0.06
    arshal
    -0.06
    POS
    -0.06
    POSITIVE LOGITS
    .pe
    0.06
    <Object
    0.06
    领导
    0.06
    Signup
    0.06
    (open
    0.06
    оці
    0.06
     interiors
    0.06
    /${
    0.06
     �
    0.06
     생산
    0.06
    Activations Density 0.002%

    No Known Activations