© 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. 89790
    Prev
    Next
    INDEX
    Explanations

    auxiliary verbs/probability

    np_max-act · gemini-2.0-flash

    This neuron spotlights hedging and advisory language—modal verbs and uncertainty markers (e.g. “it’s,” “probably,” “generally,” “okay”).

    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
     plasma
    -0.07
    前
    -0.07
     Sus
    -0.07
    contract
    -0.06
    Mag
    -0.06
     Janet
    -0.06
     peas
    -0.06
     Priest
    -0.06
    _tid
    -0.06
    ox
    -0.06
    POSITIVE LOGITS
     шляхом
    0.07
    FLAG
    0.07
    ivatel
    0.06
    uese
    0.06
     �
    0.06
    احی
    0.06
     inaccurate
    0.06
     Nich
    0.06
     če
    0.06
     norske
    0.06
    Activations Density 0.030%

    No Known Activations