© 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. 11-RESID-POST-AA
    6. 35195
    Prev
    Next
    INDEX
    Explanations

    idea

    np_max-act · gemini-2.0-flash

    This neuron primarily responds to the token "brick" (e.g., occurrences of "brick" in code or identifiers).

    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_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
    challenge
    -0.07
    颁奖
    -0.07
    돐
    -0.07
    zew
    -0.07
    .overflow
    -0.07
     surgery
    -0.07
     hiring
    -0.07
     Herr
    -0.07
     redo
    -0.07
    رار
    -0.07
    POSITIVE LOGITS
    	can
    0.07
    专用
    0.07
    高速公路
    0.06
     och
    0.06
     ô
    0.06
    宣
    0.06
    	↵
    0.06
     Fifty
    0.06
     Fant
    0.06
    饰
    0.06
    Activations Density 0.254%

    No Known Activations