© 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. 110330
    Prev
    Next
    INDEX
    Explanations

    code snippets and answers

    np_max-act · gemini-2.0-flash

    The neuron fires on the “Is the summary factually consistent with the document?” question prompt—in other words, it detects the Yes/No factual-consistency question.

    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
    _call
    -0.07
    taskId
    -0.07
     timeval
    -0.07
    wan
    -0.07
     Alb
    -0.06
    樹
    -0.06
    _serial
    -0.06
     propulsion
    -0.06
     Concord
    -0.06
    ,m
    -0.06
    POSITIVE LOGITS
    (cb
    0.06
    532
    0.06
    (commit
    0.06
     должны
    0.06
    avoid
    0.06
    .encode
    0.06
     آلة
    0.06
    undreds
    0.06
    btc
    0.06
     aud
    0.06
    Activations Density 0.011%

    No Known Activations