© 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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 11347
    Prev
    Next
    INDEX
    Explanations

    references to sewing and related crafting activities

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

    Expressions of time and past experiences

    np_acts-logits-general · gemini-2.0-flash

    past events and learning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rungsseite
    -0.76
     queſta
    -0.71
    currently
    -0.70
     Majefty
    -0.67
     Currently
    -0.67
     aktuell
    -0.66
     فريبيس
    -0.65
     actuellement
    -0.64
     huidige
    -0.64
     obecnie
    -0.64
    POSITIVE LOGITS
     damals
    0.46
     was
    0.44
     then
    0.37
    当时
    0.36
    那时
    0.36
     тогда
    0.36
    当時は
    0.35
     gave
    0.34
    那时候
    0.33
     Ver
    0.33
    Activations Density 0.539%

    No Known Activations