© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 12696
    Prev
    Next
    INDEX
    Explanations

    above

    np_max-act · gemini-2.0-flash

    This neuron fires on mid‐text discourse cues that refer back to earlier content—especially phrases like “as described above” and similar transitional markers.

    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
     tuna
    -0.07
    椅
    -0.07
     сили
    -0.07
    .descriptor
    -0.07
     lässt
    -0.06
     hva
    -0.06
    .startTime
    -0.06
    孙
    -0.06
     newbie
    -0.06
     negoci
    -0.06
    POSITIVE LOGITS
    롭
    0.07
    673
    0.06
    .getPrice
    0.06
    cae
    0.06
     آم
    0.06
    voices
    0.06
    resco
    0.06
    combine
    0.06
    ична
    0.06
    ||↵
    0.06
    Activations Density 0.009%

    No Known Activations