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. 18-GEMMASCOPE-RES-16K
    6. 1135
    Prev
    Next
    INDEX
    Explanations

    phrases and terms related to data collection and integration

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

    Combining or merging items

    np_acts-logits-general · gemini-2.0-flash

    merging and combining

    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
     يتيمه
    -0.56
    ItemLayout
    -0.53
     aload
    -0.49
     marquées
    -0.49
     dorada
    -0.48
    Impl
    -0.47
    mybatisplus
    -0.47
    ൾ
    -0.47
    참고
    -0.46
    WebServlet
    -0.45
    POSITIVE LOGITS
     together
    1.21
     disparate
    1.00
     Together
    0.94
    together
    0.92
    Together
    0.90
     TOGETHER
    0.87
     into
    0.80
     juntos
    0.76
    Into
    0.76
     Into
    0.71
    Activations Density 0.471%

    No Known Activations