© 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. 56473
    Prev
    Next
    INDEX
    Explanations

    technical documents

    np_max-act · gemini-2.0-flash

    The neuron preferentially activates on technical nouns that appear in quantitative or measurement contexts—i.e. terms occurring alongside numeric data.

    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
    	script
    -0.07
    drops
    -0.07
     Operation
    -0.07
     состав
    -0.06
     Second
    -0.06
     sensitivity
    -0.06
    ADS
    -0.06
    houses
    -0.06
    victim
    -0.06
     Regex
    -0.06
    POSITIVE LOGITS
     краще
    0.07
    xFFFFFF
    0.07
    -billion
    0.07
    .structure
    0.07
     čtyři
    0.07
    outdir
    0.07
    .Many
    0.06
    �
    0.06
    hhh
    0.06
     kde
    0.06
    Activations Density 0.310%

    No Known Activations