© 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
    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. 73762
    Prev
    Next
    INDEX
    Explanations

    Spam/Adult themed content

    np_max-act · gemini-2.0-flash

    This neuron detects multi-word spam or ad-style keyword phrases (e.g. “mail order brides,” “top 10 icon android miner,” “satellite live camera”).

    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
    ีม
    -0.07
    high
    -0.07
     هواپیم
    -0.07
    achts
    -0.06
     Lewis
    -0.06
    Pure
    -0.06
     Guest
    -0.06
     pseudo
    -0.06
    _genre
    -0.06
     CLAIM
    -0.06
    POSITIVE LOGITS
     hbox
    0.07
    _STRUCT
    0.06
     zh
    0.06
    .AL
    0.06
    _diag
    0.06
     Đo
    0.06
     Marino
    0.05
     getPage
    0.05
     SI
    0.05
     progressively
    0.05
    Activations Density 0.017%

    No Known Activations