© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 3404
    Prev
    Next
    INDEX
    Explanations

    phrases expressing honesty and transparency

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    "Honest" and its related forms

    np_acts-logits-general · gemini-2.0-flash

    speaking honestly and truthfully

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    #+#
    -0.68
     Paglinawan
    -0.66
    expandindo
    -0.65
    InjectAttribute
    -0.65
     Италијани
    -0.64
    tagHelper
    -0.57
    مصادر
    -0.55
    
    -0.54
     resultCode
    -0.52
     noDo
    -0.51
    POSITIVE LOGITS
     honest
    2.14
     honesty
    2.11
    honest
    1.90
     Honest
    1.76
    Honest
    1.75
     Honesty
    1.73
     frank
    1.64
     truthful
    1.62
     candid
    1.53
     honnête
    1.52
    Activations Density 0.246%

    No Known Activations