© 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. 8812
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    This neuron detects expressions of intent or future-oriented actions (e.g., “looking to,” “will be looking,” “to make the most,” etc.).

    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
     Batman
    -0.06
     chilly
    -0.06
     Iterate
    -0.06
     approach
    -0.06
    fieldname
    -0.06
    volume
    -0.06
     stirred
    -0.06
     cessation
    -0.06
    ând
    -0.06
     Fallon
    -0.06
    POSITIVE LOGITS
    ightly
    0.07
     kvin
    0.07
    ,h
    0.06
    ูง
    0.06
     Mehr
    0.06
     张
    0.06
    谱
    0.06
    xr
    0.06
    ≡≡
    0.06
    =<
    0.06
    Activations Density 0.043%

    No Known Activations