© 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. 37960
    Prev
    Next
    INDEX
    Explanations

    problems, negative situations

    np_max-act · gemini-2.0-flash

    The neuron activates on passive‐voice phrases in the past tense (e.g. “was flooded in”).

    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
    ica
    -0.07
    dependent
    -0.06
     Sche
    -0.06
    shed
    -0.06
     Jimmy
    -0.06
     Cure
    -0.06
    VM
    -0.06
     breadcrumb
    -0.06
     pe
    -0.06
    Shader
    -0.06
    POSITIVE LOGITS
     camb
    0.07
    _dbg
    0.06
    .validators
    0.06
    .SuppressLint
    0.06
     mơ
    0.06
     nguồn
    0.06
     MouseEventArgs
    0.06
    ～～
    0.06
    _CUSTOM
    0.06
     assms
    0.06
    Activations Density 0.243%

    No Known Activations