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. 130540
    Prev
    Next
    INDEX
    Explanations

    Repetitions and Nonsense

    np_max-act · gemini-2.0-flash

    The neuron never activates on any input tokens—it’s effectively a “dead” neuron that doesn’t detect any pattern.

    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
     Böyle
    -0.06
    	filter
    -0.06
    ثل
    -0.06
     JT
    -0.06
     ضو
    -0.06
    Howard
    -0.06
    婷
    -0.06
    otta
    -0.06
    ipay
    -0.06
    Entities
    -0.06
    POSITIVE LOGITS
     altern
    0.07
    -St
    0.07
    baseUrl
    0.07
    Comic
    0.07
    _lists
    0.07
    ลด
    0.07
     коллек
    0.06
    (lbl
    0.06
    _MA
    0.06
     될
    0.06
    Activations Density 0.080%

    No Known Activations