Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 30375
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns

    np_max-act · gemini-2.0-flash

    The neuron flags tokens that introduce biographical or career‐phase transitions—words like “After,” “His,” “retirement,” “first,” and commas that mark the start of a new life-event sentence.

    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
    =text
    -0.07
     Beirut
    -0.07
    .Measure
    -0.07
     Tecn
    -0.07
    Wrapper
    -0.07
     crafts
    -0.06
    .ali
    -0.06
    bar
    -0.06
    DAQ
    -0.06
    .IsValid
    -0.06
    POSITIVE LOGITS
    asından
    0.08
    [target
    0.06
     و
    0.06
    -days
    0.06
    μιλος
    0.06
    ös
    0.06
    押
    0.06
    цуз
    0.06
     �
    0.06
    	os
    0.06
    Activations Density 0.031%

    No Known Activations