Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 10975
    Prev
    Next
    INDEX
    Explanations

    invitations and prompts for communication

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

    Following personal communication requests

    np_acts-logits-general · gemini-2.0-flash

    private messages and emails

    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
     Савезне
    -0.72
     createSprite
    -0.71
    SceneManagement
    -0.70
    SourceChecksum
    -0.70
    Попис
    -0.65
    aarrggbb
    -0.63
     podjela
    -0.62
     jLabel
    -0.62
    bootstrapcdn
    -0.61
    psies
    -0.60
    POSITIVE LOGITS
     PM
    1.27
     email
    1.24
     inbox
    1.10
    PM
    1.08
     private
    1.07
     pm
    1.05
     messaging
    1.04
     contact
    1.03
     DM
    1.02
     DMs
    0.97
    Activations Density 0.125%

    No Known Activations