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. 14394
    Prev
    Next
    INDEX
    Explanations

    phrases related to hanging or being suspended

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

    Forms of the word "hang"

    np_acts-logits-general · gemini-2.0-flash

    hangs, hanging, and related verbs

    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
    otas
    -0.51
    tiera
    -0.48
     pisos
    -0.45
    Contributor
    -0.44
    ľad
    -0.44
    RegressionTest
    -0.43
    localctx
    -0.41
     civ
    -0.41
     Steele
    -0.41
     loci
    -0.41
    POSITIVE LOGITS
     HANG
    1.44
     hang
    1.40
     Hang
    1.37
     hung
    1.33
     hanging
    1.29
     hangs
    1.29
     Hanging
    1.29
    Hanging
    1.27
    Hang
    1.24
    hanging
    1.18
    Activations Density 0.080%

    No Known Activations