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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 22-GEMMASCOPE-RES-16K
    6. 10951
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the need for emphasis or attention in discussing various topics

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

    "Attention has been" or "attention is/should be"

    np_acts-logits-general · gemini-2.0-flash

    attention is focused or given

    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
    Билгалдахарш
    -0.74
    دانشنامهٔ
    -0.62
     Behold
    -0.62
     Affecting
    -0.61
     SafeMath
    -0.61
    ffes
    -0.60
    findpost
    -0.60
    Вз
    -0.59
    happen
    -0.59
    doir
    -0.58
    POSITIVE LOGITS
     placed
    0.63
    placed
    0.50
     trajets
    0.48
    elemField
    0.46
     Placed
    0.46
     accorded
    0.46
    Placed
    0.46
     contextLoads
    0.45
     given
    0.45
    bitField
    0.42
    Activations Density 0.345%

    No Known Activations