© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 74687
    Prev
    Next
    INDEX
    Explanations

    Arranging objects

    np_max-act · gemini-2.0-flash

    instructions or steps on how to stack items in a stable manner.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron activates on tokens in user requests or assistant replies that give or ask for step-by-step “stacking” instructions (e.g. “stack,” “stable,” “manner,” etc.).

    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
    /lo
    -0.07
     sinister
    -0.06
    lene
    -0.06
     inhibition
    -0.06
     hiring
    -0.06
     whose
    -0.06
     realizado
    -0.06
     Payload
    -0.06
    αρ
    -0.06
     avenues
    -0.06
    POSITIVE LOGITS
     rağmen
    0.07
     j
    0.07
    ycopg
    0.07
    _HAND
    0.07
    URIComponent
    0.07
    (ag
    0.06
    _COM
    0.06
    _FW
    0.06
    кувати
    0.06
    lli
    0.06
    Activations Density 0.011%

    No Known Activations