© 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
    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. 126197
    Prev
    Next
    INDEX
    Explanations

    varied content

    np_max-act · gemini-2.0-flash

    This neuron detects informal conversational discourse markers or filler words (e.g., “okay,” “so,” “who better,” “straight to”), flagging a colloquial or spoken-tone transition in the text.

    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
     Got
    -0.07
    .Getter
    -0.06
    زا
    -0.06
    Ot
    -0.06
    ileged
    -0.06
    ⠀⠀
    -0.06
    _take
    -0.06
     GOT
    -0.06
    章
    -0.06
     deer
    -0.06
    POSITIVE LOGITS
    (Byte
    0.07
     tấm
    0.07
    	ex
    0.07
     Relay
    0.07
    .last
    0.07
    imentary
    0.07
     uint
    0.06
    cil
    0.06
    .setStroke
    0.06
    BufferSize
    0.06
    Activations Density 0.045%

    No Known Activations