© 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. 11-RESID-POST-AA
    11. 86887
    Prev
    Next
    INDEX
    Explanations

    Javascript

    np_max-act · gemini-2.0-flash

    alphanumeric substrings that look like long resource-integrity or hash-like tokens (long sequences of letters/digits).

    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
     on
    -0.07
    ,True
    -0.07
    ersed
    -0.07
    (Order
    -0.07
    apphire
    -0.07
    PASS
    -0.07
    編
    -0.07
     pour
    -0.07
     in
    -0.07
    .POS
    -0.06
    POSITIVE LOGITS
    עוב
    0.08
    `t
    0.08
    t
    0.07
     multit
    0.07
    ht
    0.07
    dat
    0.07
    utenant
    0.07
    bart
    0.07
    mt
    0.07
    ᴛ
    0.07
    Activations Density 1.518%

    No Known Activations