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. 24-GEMMASCOPE-RES-16K
    6. 4318
    Prev
    Next
    INDEX
    Explanations

    technical terms and metrics related to radio frequency surveys and astrophysical observations

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

    galaxies, halos, and redshift

    np_acts-logits-general · gemini-2.0-flash

    astronomical galaxies and their properties

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Geplaatst
    -0.84
     typelib
    -0.80
     متعلقه
    -0.77
     utafitiHapana
    -0.73
     defaultstate
    -0.70
    脚注の使い方
    -0.67
    FormTagHelper
    -0.63
    styleType
    -0.62
     >=",
    -0.62
    RectangleBorder
    -0.61
    POSITIVE LOGITS
     galaxies
    0.57
     galaxy
    0.56
    #+#
    0.54
     halo
    0.54
     halos
    0.53
     GOODS
    0.50
    halo
    0.49
     cluster
    0.44
     Sal
    0.44
     Local
    0.44
    Activations Density 0.568%

    No Known Activations