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
    4. Residual Stream - 16k
    5. 37-GEMMASCOPE-RES-16K
    6. 5799
    Prev
    Next
    INDEX
    Explanations

    technical terminology related to data structures and programming concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.37.hook_resid_post
    Hook Layer
    37
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Portail
    -0.54
     ModelExpression
    -0.49
    zydent
    -0.46
     مشين
    -0.45
    ScopeManager
    -0.45
     PeEnEo
    -0.42
     <>",
    -0.40
    Nuorodos
    -0.40
     uVar
    -0.40
     <=",
    -0.39
    POSITIVE LOGITS
     row
    3.28
     rows
    3.05
     Row
    2.88
     Rows
    2.69
    Row
    2.67
     ROW
    2.56
    row
    2.53
    Rows
    2.36
    rows
    2.30
    ROW
    2.22
    Activations Density 0.946%

    No Known Activations