© 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. 89825
    Prev
    Next
    INDEX
    Explanations

    true

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that assert truth or logical validity (e.g. “true,” “holds,” “implies”).

    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
    láš
    -0.08
    ẵng
    -0.07
    slaught
    -0.06
    .byId
    -0.06
    alez
    -0.06
     michael
    -0.06
    filepath
    -0.06
    .firebaseio
    -0.06
     newcomers
    -0.06
     awhile
    -0.06
    POSITIVE LOGITS
    avadoc
    0.07
    _JS
    0.06
     نشر
    0.06
    ΟΥ
    0.06
    ��
    0.06
     CRE
    0.06
    TN
    0.06
    =target
    0.06
     fond
    0.06
     vari
    0.06
    Activations Density 0.011%

    No Known Activations