© 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. 22-GEMMASCOPE-RES-16K
    6. 3528
    Prev
    Next
    INDEX
    Explanations

    references to advanced imaging or scanning technologies

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

    "three" or "3D"

    np_acts-logits-general · gemini-2.0-flash

    3D, three-dimensional, depth

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    存于互联网档案馆
    -0.63
     pinulongan
    -0.55
     nifer
    -0.46
    WriteAttribute
    -0.46
     eleventh
    -0.45
     otomatig
    -0.45
     متعلقه
    -0.43
    PerformLayout
    -0.42
     tendre
    -0.41
     unqualified
    -0.40
    POSITIVE LOGITS
     three
    1.07
     Three
    1.05
     THREE
    1.02
     ثلاث
    0.98
     трех
    0.98
     tiga
    0.96
    3
    0.95
     virtual
    0.91
     трёх
    0.91
    three
    0.90
    Activations Density 0.584%

    No Known Activations