© 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. Qwen2.5-7B-IT
    4. Resid Post - 131k
    5. 15-RESID-POST-AA
    6. 41945
    Prev
    Next
    INDEX
    Explanations

    Sentence/phrase beginnings

    np_max-act · gemini-2.0-flash

    The neuron detects sentence- or paragraph-initial tokens (short, capitalized function words like "En", "In", "So", "Este") that mark the start of a new sentence or section.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cena
    -0.07
    <iostream
    -0.07
     converse
    -0.07
     licked
    -0.06
    က
    -0.06
     SPORT
    -0.06
    _INET
    -0.06
    蓁
    -0.06
    sin
    -0.06
     Lista
    -0.06
    POSITIVE LOGITS
    emption
    0.07
    倒闭
    0.07
     reactor
    0.07
    ImageData
    0.07
     skeletal
    0.07
    ifferent
    0.07
    红外
    0.07
    relevant
    0.07
    ventional
    0.07
    Operation
    0.06
    Activations Density 0.152%

    No Known Activations