© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 14673
    Prev
    Next
    INDEX
    Explanations

    matics, maximum, miscellaneous

    np_max-act · gemini-2.0-flash

    long, formal technical vocabulary and configuration identifiers—especially abstract nouns/adjectives with Latinate suffixes (e.g., -tion/-sion, -ary/-ory, -ous) and similar polysyllabic terms.

    oai_token-act-pair · gpt-5Triggered 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
    _true
    -0.08
    lica
    -0.07
    พร
    -0.07
     SSR
    -0.07
    tte
    -0.06
     Foods
    -0.06
     trời
    -0.06
    .Popup
    -0.06
    -chain
    -0.06
    /modules
    -0.06
    POSITIVE LOGITS
    DirectoryName
    0.07
     envelope
    0.07
    örper
    0.07
     шаг
    0.07
     advantageous
    0.07
     cabel
    0.07
     signIn
    0.07
    味道
    0.07
    吸引力
    0.06
    闪光
    0.06
    Activations Density 0.285%

    No Known Activations