© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 91682
    Prev
    Next
    INDEX
    Explanations

    Chemical reaction

    np_max-act · gemini-2.0-flash

    The neuron fires on verbs that describe chemical‐process steps (e.g. “formation,” “results,” “reaction”) in synthesis write-ups.

    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
     similarities
    -0.07
     Cond
    -0.07
     anal
    -0.06
     Стар
    -0.06
     Î
    -0.06
     swallowing
    -0.06
    ixture
    -0.06
    Closed
    -0.06
    .TextChanged
    -0.06
    Port
    -0.06
    POSITIVE LOGITS
    :center
    0.07
    orientation
    0.07
    _BEGIN
    0.06
     disproportion
    0.06
     ruce
    0.06
    	instance
    0.06
    
    0.06
     어디
    0.06
     dolor
    0.06
     userid
    0.06
    Activations Density 0.014%

    No Known Activations