© 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. 657
    Prev
    Next
    INDEX
    Explanations

    Estimated versus true figures

    np_max-act · gemini-2.0-flash

    This neuron fires on in‐text numerical statistics—particularly decimal numbers and data figures—highlighting places where precise metrics or measurements appear.

    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
    gettext
    -0.06
     Typed
    -0.06
    aci
    -0.06
     branded
    -0.06
    erte
    -0.06
    纸
    -0.06
     strlen
    -0.06
    ेब
    -0.06
    аци
    -0.06
    aque
    -0.06
    POSITIVE LOGITS
    .j
    0.08
    -holder
    0.07
     аж
    0.07
    (fragment
    0.06
    (Client
    0.06
    .frequency
    0.06
     двиг
    0.06
     Lies
    0.06
    bou
    0.06
    샵
    0.06
    Activations Density 0.048%

    No Known Activations