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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Qwen2.5-7B-IT
    4. Resid Post - 131k
    5. 19-RESID-POST-AA
    6. 36268
    Prev
    Next
    INDEX
    Explanations

    general topics

    np_max-act-logits · gemini-2.0-flash

    concrete nouns and key informational content words that carry semantic weight in the text.

    oai_token-act-pair · claude-4-5-haikuTriggered by @xiaoqingsun004
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_19/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.19.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    script
    -0.07
    flush
    -0.07
     BufferedReader
    -0.06
     measles
    -0.06
    begin
    -0.06
    (enabled
    -0.06
     throne
    -0.06
    (border
    -0.06
     election
    -0.06
    ensure
    -0.06
    POSITIVE LOGITS
    村落
    0.08
    🔹
    0.07
    教学质量
    0.07
    ptides
    0.07
    使劲
    0.06
     narr
    0.06
    街区
    0.06
    ř
    0.06
    杼
    0.06
    Ị
    0.06
    Activations Density 2.402%

    No Known Activations