© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 2508
    Prev
    Next
    INDEX
    Explanations

    terms related to oversight and influence in decision-making contexts

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

    Appears before or contains "v"

    np_acts-logits-general · gemini-2.0-flash

    Riv convention Dave doubling

    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
    GEBURTSDATUM
    -0.80
    er
    -0.71
    createCanvas
    -0.65
    ね
    -0.62
     fight
    -0.58
    cipar
    -0.57
    UpInside
    -0.56
    ngrx
    -0.56
    fight
    -0.55
    eric
    -0.54
    POSITIVE LOGITS
     Merv
    0.90
     mav
    0.86
     bv
    0.82
     tev
    0.82
     xv
    0.82
     Hv
    0.82
     Mv
    0.81
     WV
    0.80
     MAV
    0.79
     Jove
    0.78
    Activations Density 0.665%

    No Known Activations