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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 8161
    Prev
    Next
    INDEX
    Explanations

    future tense, continuation (edited)

    np_max-act · gemini-2.0-flash

    The neuron responds to language laying out future or ongoing research plans (e.g. “will continue,” “studies will,” “will be performed”).

    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
    astreet
    -0.06
     Arena
    -0.06
    imb
    -0.06
     Ky
    -0.06
     caves
    -0.06
    .str
    -0.06
     MAC
    -0.06
     замет
    -0.06
    ...\
    -0.06
     dh
    -0.05
    POSITIVE LOGITS
    marginLeft
    0.07
     vzpom
    0.07
    -rule
    0.07
    .current
    0.07
    -disabled
    0.06
     authDomain
    0.06
    颜
    0.06
    <Comment
    0.06
    conomics
    0.06
     customized
    0.06
    Activations Density 0.035%

    No Known Activations