© 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. 6549
    Prev
    Next
    INDEX
    Explanations

    else

    np_max-act · gemini-2.0-flash

    The neuron activates on modal or future‐tense auxiliaries (e.g. “would,” “might,” “will”) that signal potential or conditional outcomes.

    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
    COVID
    -0.08
    _F
    -0.07
     داو
    -0.06
    ردد
    -0.06
    ingga
    -0.06
    �
    -0.06
    (volume
    -0.06
     VM
    -0.06
    swagger
    -0.06
    ihad
    -0.06
    POSITIVE LOGITS
    ηση
    0.06
    lated
    0.06
     ind
    0.06
     ansch
    0.06
     <*
    0.06
     ult
    0.06
    lehem
    0.06
     전에
    0.06
    _();↵
    0.06
    tag
    0.06
    Activations Density 0.015%

    No Known Activations