© 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. 40-GEMMASCOPE-RES-16K
    6. 2032
    Prev
    Next
    INDEX
    Explanations

    quantifiable metrics or statistics related to various subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 40-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_40/width_16k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.40.hook_resid_post
    Hook Layer
    40
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pleaſure
    -0.58
     AssemblyCompany
    -0.47
    二十
    -0.45
     JpaRepository
    -0.44
    Parcelize
    -0.43
    ſelf
    -0.42
     itſelf
    -0.41
    сылкі
    -0.41
    十几
    -0.41
     zehn
    -0.40
    POSITIVE LOGITS
     thousands
    1.16
     thousand
    1.06
    thousands
    1.01
     Thousands
    0.98
    Thousands
    0.98
    三千
    0.94
     THOUS
    0.89
    thousand
    0.87
     Thousand
    0.84
    Thousand
    0.78
    Activations Density 0.680%

    No Known Activations