© 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. 6910
    Prev
    Next
    INDEX
    Explanations

    blog posts

    np_max-act · gemini-2.0-flash

    The neuron detects references to embedded photos or image captions (e.g. “picture,” “shot,” “pics,” “here is,” “finished”).

    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
    Objects
    -0.08
    xing
    -0.08
    akk
    -0.07
     merciless
    -0.07
     []*
    -0.07
     trouble
    -0.06
    -gallery
    -0.06
     operated
    -0.06
     Costume
    -0.06
    FileInfo
    -0.06
    POSITIVE LOGITS
    wiąz
    0.07
    /Branch
    0.07
    smouth
    0.06
     sàng
    0.06
     trộn
    0.06
    	align
    0.06
    -radio
    0.06
     espacio
    0.06
    bsp
    0.06
    ?>">↵
    0.06
    Activations Density 0.018%

    No Known Activations