© 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
    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. 41578
    Prev
    Next
    INDEX
    Explanations

    strength

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of a company’s “key strengths” or core capabilities in corporate introductions.

    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
     مسائل
    -0.07
    _dict
    -0.06
     onPostExecute
    -0.06
     incid
    -0.06
    _code
    -0.06
     apopt
    -0.06
     ApplicationRecord
    -0.06
     XmlNode
    -0.06
    .Glide
    -0.06
     دقیقه
    -0.06
    POSITIVE LOGITS
     strengths
    0.10
     weaknesses
    0.08
    rypton
    0.07
    十八
    0.07
     talented
    0.07
     ked
    0.07
     weakness
    0.06
     ung
    0.06
     team
    0.06
    atat
    0.06
    Activations Density 0.016%

    No Known Activations