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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 11078
    Prev
    Next
    INDEX
    Explanations

    comparative data on medical patient outcomes and test results

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

    preceding "compared" or "than"

    np_acts-logits-general · gemini-2.0-flash

    compared or versus

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    principalTable
    -0.77
    #+#
    -0.69
     propOrder
    -0.69
    setVerticalGroup
    -0.68
     createState
    -0.68
    AutoScale
    -0.66
    abestanden
    -0.63
     pinulongan
    -0.60
    LookAnd
    -0.60
    */].
    -0.60
    POSITIVE LOGITS
     compared
    4.06
    compared
    3.40
     Compared
    3.03
    Compared
    2.85
     compare
    2.76
     versus
    2.76
     vs
    2.73
     dibandingkan
    2.64
     comparison
    2.53
     compares
    2.51
    Activations Density 1.628%

    No Known Activations