© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 28-GEMMASCOPE-RES-16K
    11. 924
    Prev
    Next
    INDEX
    Explanations

    descriptive terms and identifiers related to various subjects, including names and features

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    "... as" phrases, often definitions or descriptions

    np_acts-logits-general · gemini-2.0-flash

    as or as a

    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
    Negative Logits
     noDo
    -0.64
    SequentialGroup
    -0.54
    fromnode
    -0.50
     &___
    -0.50
    transQ
    -0.48
    */].
    -0.47
    AddTagHelper
    -0.47
    GOTREF
    -0.46
    cydow
    -0.45
    AutoScale
    -0.45
    POSITIVE LOGITS
     as
    1.72
     sebagai
    1.47
     ως
    1.10
    sebagai
    1.06
     Sebagai
    1.05
     jako
    0.90
    作为
    0.89
    Sebagai
    0.84
     как
    0.84
     als
    0.83
    Activations Density 0.806%

    No Known Activations