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

    Neuronpedia

    Jacobian LensNEW
    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. 38856
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively activates on numeric constants—especially floating-point literals with decimal points—in code.

    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
     ridge
    -0.07
    (Member
    -0.07
    緊
    -0.07
     зараз
    -0.06
    Exists
    -0.06
     birinci
    -0.06
    _VAL
    -0.06
     velice
    -0.06
     накоп
    -0.06
     towel
    -0.06
    POSITIVE LOGITS
     Blasio
    0.07
     contest
    0.06
    Ground
    0.06
     ostatní
    0.06
    เสนอ
    0.06
    _Enable
    0.06
    ::|
    0.06
    _answer
    0.06
    .y
    0.06
     authorize
    0.06
    Activations Density 0.019%

    No Known Activations