Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1542
    Prev
    Next
    INDEX
    Explanations

    terms related to moving and relocation events

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

    relocation or changing residence

    np_acts-logits-general · gemini-2.0-flash

    moving and relocation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     Waray
    -0.54
     дописавши
    -0.48
     onError
    -0.47
    ReadKey
    -0.47
    SuccessListener
    -0.45
    GUILayout
    -0.44
    icure
    -0.42
    protoimpl
    -0.42
    inate
    -0.42
    mentary
    -0.42
    POSITIVE LOGITS
     relocation
    2.40
     move
    2.31
     relocate
    2.29
     relocating
    2.22
     moved
    2.17
     relocated
    2.13
     Move
    2.11
    move
    2.08
     Relocation
    2.06
     moving
    2.06
    Activations Density 0.341%

    No Known Activations