© 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. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 68747
    Prev
    Next
    INDEX
    Explanations

    hangman game

    np_max-act · gemini-2.0-flash

    The neuron activates when the text refers to the “hangman” game (i.e. occurrences of “hang” and “man” together).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-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
     cry
    -0.06
     manuals
    -0.06
    uos
    -0.06
     Clock
    -0.06
     cake
    -0.06
     HttpMethod
    -0.06
     giữa
    -0.06
    .completed
    -0.06
     `%
    -0.06
    กต
    -0.06
    POSITIVE LOGITS
     Lantern
    0.07
    タイ
    0.07
    built
    0.07
     NUMBER
    0.06
     Ре
    0.06
     कब
    0.06
     bm
    0.06
    labilir
    0.06
     Cd
    0.06
     án
    0.06
    Activations Density 0.006%

    No Known Activations