© 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. 51921
    Prev
    Next
    INDEX
    Explanations

    instructions

    np_max-act · gemini-2.0-flash

    This neuron detects the markers and verbs that introduce numbered steps in instructional or procedural lists.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    steps related to developing interactive applications.

    oai_token-act-pair · gpt-4o-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
    [A
    -0.06
    _Key
    -0.06
     corpus
    -0.06
    .sp
    -0.06
     sert
    -0.06
     action
    -0.06
    Bed
    -0.06
    ุญ
    -0.06
     troub
    -0.06
     Dice
    -0.05
    POSITIVE LOGITS
    ernity
    0.07
     Establishment
    0.06
    akest
    0.06
     комнат
    0.06
    (-
    0.06
     invites
    0.06
    ">×</
    0.06
     dataType
    0.06
    ản
    0.06
     Matter
    0.06
    Activations Density 0.077%

    No Known Activations