© 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. 89687
    Prev
    Next
    INDEX
    Explanations

    fugitives hiding

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens (e.g. years, dates, and other digit sequences).

    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
    pulse
    -0.07
    icao
    -0.06
     ayn
    -0.06
     AK
    -0.06
     vest
    -0.06
    SPORT
    -0.06
     bots
    -0.06
     INPUT
    -0.06
    .seek
    -0.06
    .lists
    -0.06
    POSITIVE LOGITS
     getValue
    0.06
     proposes
    0.06
    StartupScript
    0.06
     pprint
    0.06
     حدود
    0.06
    EObject
    0.06
    -hide
    0.06
     coherent
    0.06
    progress
    0.06
    /package
    0.06
    Activations Density 0.022%

    No Known Activations