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. 18-GEMMASCOPE-RES-16K
    6. 15208
    Prev
    Next
    INDEX
    Explanations

    the word "through" and its variations in various contexts related to education and processes

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

    "through" or "Through"

    np_acts-logits-general · gemini-2.0-flash

    throughfall and through the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    RenderAtEndOf
    -0.47
     frontale
    -0.47
     Púb
    -0.46
     marins
    -0.45
    <em>
    -0.43
    vanje
    -0.43
     autoridad
    -0.42
    WebServlet
    -0.41
     Reiz
    -0.40
    new
    -0.40
    POSITIVE LOGITS
    THROUGH
    1.61
    through
    1.54
    Through
    1.45
     Through
    1.41
     THROUGH
    1.39
     through
    1.38
    tagHelperRunner
    1.13
     hindurch
    1.12
     Thru
    1.11
     thru
    1.11
    Activations Density 0.125%

    No Known Activations