Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 84590
    Prev
    Next
    INDEX
    Explanations

    man

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of "man" (as in manual pages) or references to consulting the system manual.

    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
     Vulcan
    -0.06
     후기
    -0.06
     тро
    -0.06
     Detroit
    -0.06
     Durant
    -0.06
    _epi
    -0.06
     ن
    -0.06
    (dispatch
    -0.06
     regist
    -0.06
    oscopic
    -0.06
    POSITIVE LOGITS
    ните
    0.07
     vowels
    0.06
    _CYCLE
    0.06
     différentes
    0.06
     algum
    0.06
    동안
    0.06
    ――――
    0.06
     Wh
    0.06
     SOUTH
    0.06
    intersection
    0.06
    Activations Density 0.004%

    No Known Activations