© 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. 92279
    Prev
    Next
    INDEX
    Explanations

    code and system errors

    np_max-act · gemini-2.0-flash

    The neuron strongly responds to the model’s end-of-text marker tokens (e.g. “<|eot_id|>”).

    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
     TimeZone
    -0.07
     mejor
    -0.06
    GLIGENCE
    -0.06
    )
    ↵
    ↵
    -0.06
    nton
    -0.06
     proces
    -0.06
     пояс
    -0.06
    toolbox
    -0.06
     pol
    -0.06
    .rank
    -0.06
    POSITIVE LOGITS
    _death
    0.07
    States
    0.07
     denial
    0.07
     launching
    0.06
    Def
    0.06
    Define
    0.06
    <File
    0.06
    Projection
    0.06
    napshot
    0.06
    始
    0.06
    Activations Density 0.002%

    No Known Activations