© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 7041
    Prev
    Next
    INDEX
    Explanations

    collaborative or collective efforts in sharing knowledge and expertise

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

    Sharing knowledge or expertise

    np_acts-logits-general · gemini-2.0-flash

    sharing expertise and knowledge

    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
    cellation
    -0.63
    Obrázky
    -0.59
    ########.
    -0.57
    AndEndTag
    -0.56
     ái
    -0.53
     podio
    -0.53
    //
    -0.52
    禧
    -0.52
     henvis
    -0.51
     tolerably
    -0.51
    POSITIVE LOGITS
     expertise
    1.57
    expertise
    1.40
     wisdom
    1.26
     Expertise
    1.22
    Expertise
    1.13
     expert
    1.08
     knowledge
    1.03
     EXPERT
    1.02
     sagesse
    1.00
    wisdom
    1.00
    Activations Density 0.187%

    No Known Activations