© 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. 28-GEMMASCOPE-RES-16K
    6. 7664
    Prev
    Next
    INDEX
    Explanations

    elements related to data structure fields and types

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

    Followed by parentheses or numbers

    np_acts-logits-general · gemini-2.0-flash

    identifiers then delimiters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aults
    -0.54
    ðsíða
    -0.54
    
    -0.52
     laſſen
    -0.52
    ceptre
    -0.52
     zuſammen
    -0.52
    ſſung
    -0.51
    mybatisplus
    -0.50
    iſen
    -0.50
    tisserie
    -0.49
    POSITIVE LOGITS
    withstanding
    0.57
    которые
    0.47
    Personendaten
    0.42
    );
    0.39
    )
    0.37
    )(
    0.36
    ).
    0.36
    ),
    0.36
    üyada
    0.36
    ):
    0.35
    Activations Density 1.035%

    No Known Activations