© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 92108
    Prev
    Next
    INDEX
    Explanations

    for/to

    np_max-act · gemini-2.0-flash

    This neuron detects definitional phrases of the form “X is the name for…” (e.g. the pattern “name … for”).

    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
     employing
    -0.07
    墨
    -0.07
     �
    -0.06
    佐
    -0.06
     employed
    -0.06
     {{$
    -0.06
    тах
    -0.06
    -consuming
    -0.06
    اها
    -0.06
    Ok
    -0.06
    POSITIVE LOGITS
    _='
    0.07
     disdain
    0.06
    ()>
    0.06
     partition
    0.06
     felon
    0.06
     electrodes
    0.06
     alias
    0.06
    ItemAt
    0.06
     prompting
    0.06
    pletely
    0.06
    Activations Density 0.036%

    No Known Activations