Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 7210
    Prev
    Next
    INDEX
    Explanations

    special characters and formatting elements in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    twimg
    -0.62
     parachoque
    -0.56
     termica
    -0.54
    onAttach
    -0.52
    pecabe
    -0.52
    verifyException
    -0.52
    RepeatedField
    -0.51
     læng
    -0.51
    chartInstance
    -0.49
    󠁿
    -0.49
    POSITIVE LOGITS
     CYTHON
    0.35
    fontawesome
    0.34
     capac
    0.34
     positive
    0.32
    WebElementEntity
    0.32
    __*/
    0.31
    tvguidetime
    0.30
    AddWithValue
    0.30
    Dabei
    0.30
     mạnh
    0.29
    Activations Density 0.079%

    No Known Activations