© 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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10357
    Prev
    Next
    INDEX
    Explanations

    phrases related to improvement and progress in performance

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

    Text following "<start_of_turn>user"

    np_acts-logits-general · gemini-2.0-flash

    shaky defensively

    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
     léka
    -0.38
     váš
    -0.34
    RequestMapping
    -0.34
     Pantalla
    -0.33
     जीवन
    -0.33
     engraçadas
    -0.33
    fvar
    -0.33
    话
    -0.32
     frattempo
    -0.32
     life
    -0.32
    POSITIVE LOGITS
     AssemblyTitle
    0.71
    addCriterion
    0.57
     '\\;'
    0.55
    يميديا
    0.53
     jScrollPane
    0.51
    findpost
    0.50
     executes
    0.49
     <<<<<<<<<<<<<<
    0.49
     HFILL
    0.49
    Numerology
    0.49
    Activations Density 0.093%

    No Known Activations