© 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. 108789
    Prev
    Next
    INDEX
    Explanations

    inference

    np_max-act · gemini-2.0-flash

    The neuron activates on words about type‐parameter inference or template/type deduction (e.g. “infers,” “inferred,” “deduction,” “deduced”).

    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
     mish
    -0.07
     paragraph
    -0.06
    frame
    -0.06
    vergence
    -0.06
     whisper
    -0.06
    raham
    -0.06
    payload
    -0.06
     spacecraft
    -0.06
    -details
    -0.06
     Numbers
    -0.06
    POSITIVE LOGITS
    .BLUE
    0.07
     ауд
    0.07
     sidewalk
    0.06
    .getPosition
    0.06
     İşte
    0.06
    .ByteString
    0.06
    month
    0.06
     okolí
    0.06
     北
    0.06
    (lo
    0.06
    Activations Density 0.006%

    No Known Activations