© 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. 13459
    Prev
    Next
    INDEX
    Explanations

    is

    np_max-act · gemini-2.0-flash

    This neuron detects the assistant’s phrasing when it introduces a proposed outline or sample—e.g. “Based on the information…,” “Here’s a…,” signaling the start of a structured suggestion or plan.

    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
     میشود
    -0.06
    entarios
    -0.06
     sed
    -0.06
    445
    -0.06
     "="
    -0.05
    .med
    -0.05
    vergence
    -0.05
    cams
    -0.05
    scene
    -0.05
    731
    -0.05
    POSITIVE LOGITS
    Restricted
    0.07
    storybook
    0.07
    .writeValue
    0.07
    eness
    0.07
    шую
    0.07
    仪
    0.07
     separating
    0.06
    COMPLETE
    0.06
     transferred
    0.06
     Swamp
    0.06
    Activations Density 0.052%

    No Known Activations