© 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. 4330
    Prev
    Next
    INDEX
    Explanations

    references to sources or origin points of information

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

    "source" or "Source"

    np_acts-logits-general · gemini-2.0-flash

    source followed by colon or file/code

    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
     pestaña
    -0.60
     ProgressDialog
    -0.60
    Vidite
    -0.56
     onItemClick
    -0.54
     tslint
    -0.52
    getInteger
    -0.52
    findOneAndUpdate
    -0.52
     proprement
    -0.50
    ements
    -0.49
     ajedrez
    -0.49
    POSITIVE LOGITS
     source
    1.48
     Source
    1.44
    source
    1.43
    Source
    1.35
     SOURCE
    1.33
    SOURCE
    1.15
     sources
    1.12
     Sources
    1.09
    ource
    1.05
    sources
    1.05
    Activations Density 0.104%

    No Known Activations