© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 10-GEMMASCOPE-RES-16K
    6. 15529
    Prev
    Next
    INDEX
    Explanations

    code or markup style elements and structural document components

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AddTagHelper
    -0.91
    awtextra
    -0.83
    rungsseite
    -0.79
     dAtA
    -0.74
    <unused79>
    -0.71
    <pad>
    -0.71
    <unused23>
    -0.71
    <unused41>
    -0.71
    <unused8>
    -0.71
    [@BOS@]
    -0.71
    POSITIVE LOGITS
    ↵↵
    0.36
    <h2>
    0.33
    <eos>
    0.33
    ↵↵↵↵
    0.31
    ↵↵↵
    0.31
    ↵↵↵↵↵↵↵↵
    0.31
    ↵↵↵↵↵↵↵
    0.29
    ↵↵↵↵↵
    0.28
     caza
    0.28
    ↵
    0.28
    Activations Density 0.374%

    No Known Activations