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

    Neuronpedia

    APICircuit 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. 104198
    Prev
    Next
    INDEX
    Explanations

    by/from

    np_max-act · gemini-2.0-flash

    The neuron specializes in spotting passive‐voice descriptions (past‐participle constructions) that mark how or when something was made, released, or made available.

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

    information about virtual reality devices.

    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
     ville
    -0.08
     pořad
    -0.06
    zeň
    -0.06
    Azure
    -0.06
    aç
    -0.06
    vail
    -0.06
    metal
    -0.06
     dị
    -0.06
    _wo
    -0.06
    ет
    -0.06
    POSITIVE LOGITS
    (GL
    0.07
    oscope
    0.07
    -encoded
    0.06
    .IntPtr
    0.06
    -parent
    0.06
    clinical
    0.06
    -approved
    0.06
     небольш
    0.06
    _ASSERT
    0.06
     _
    ↵
    0.06
    Activations Density 0.062%

    No Known Activations