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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14485
    Prev
    Next
    INDEX
    Explanations

    instances of high numerical values or significant numerical data within a context

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

    Follows a bullet point, asterisk, or separator

    np_acts-logits-general · gemini-2.0-flash

    lists of items

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    <unused41>
    -1.23
    <unused74>
    -1.22
    <unused43>
    -1.22
    <unused52>
    -1.22
    <unused14>
    -1.21
    <unused8>
    -1.21
    <unused16>
    -1.21
    <unused17>
    -1.21
    <pad>
    -1.21
    [@BOS@]
    -1.21
    POSITIVE LOGITS
    ↵↵
    0.47
    The
    0.44
    Of
    0.43
    For
    0.43
    1
    0.43
    This
    0.40
    I
    0.40
    $\
    0.40
    In
    0.39
    A
    0.39
    Activations Density 0.051%

    No Known Activations