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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 68278
    Prev
    Next
    INDEX
    Explanations

    technical documentation/forum posts

    np_max-act · gemini-2.0-flash

    This neuron flags tokens characteristic of technical or code-like contexts—such as parentheses, command-line switches, code identifiers (e.g. “ref”, “log”), and numeric or special symbols.

    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
     wield
    -0.07
     incomplete
    -0.07
    根本
    -0.06
    sent
    -0.06
     Mezi
    -0.06
     máy
    -0.06
     ولد
    -0.06
     ornament
    -0.06
    _pin
    -0.06
    Expired
    -0.06
    POSITIVE LOGITS
    *******
    ↵
    0.07
    urers
    0.06
     Turing
    0.06
    над
    0.06
    栏
    0.06
    .Change
    0.06
    uttgart
    0.06
    iciencies
    0.06
    ネル
    0.06
    кан
    0.06
    Activations Density 0.090%

    No Known Activations