© 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. 42544
    Prev
    Next
    INDEX
    Explanations

    chemical synthesis

    np_max-act · gemini-2.0-flash

    This neuron specifically detects the special “<|start_header_id|>” header‐ID marker token in the text.

    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
     MU
    -0.07
    oyal
    -0.06
     Hast
    -0.06
     Bry
    -0.06
     translator
    -0.06
    \Type
    -0.06
     swipe
    -0.06
     affluent
    -0.06
     情
    -0.06
     Name
    -0.06
    POSITIVE LOGITS
     stationed
    0.06
     Batch
    0.06
     sơn
    0.06
    设备
    0.06
     zun
    0.06
     мер
    0.06
    ,value
    0.06
     birisi
    0.06
     Equals
    0.06
     založ
    0.06
    Activations Density 0.047%

    No Known Activations