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. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 130762
    Prev
    Next
    INDEX
    Explanations

    phen and yl

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the chemical root “phenyl” (as in “phenyl-…” in compound names).

    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
     novels
    -0.07
     MT
    -0.06
     LEVEL
    -0.06
    Clip
    -0.06
     переда
    -0.06
     archives
    -0.06
     VIDEO
    -0.06
    Vars
    -0.06
    يفة
    -0.06
    boards
    -0.06
    POSITIVE LOGITS
    /div
    0.07
     subsequent
    0.07
    eref
    0.06
    П
    0.06
    signed
    0.06
    лик
    0.06
                    	
    0.06
    /Error
    0.06
     ofApp
    0.06
    div
    0.06
    Activations Density 0.025%

    No Known Activations