© 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. 31-GEMMASCOPE-RES-16K
    6. 9681
    Prev
    Next
    INDEX
    Explanations

    coding-related commands and functions

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

    code structure and syntax elements like method definitions, assignments, and Ruby/programming language constructs.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     wsz
    -0.41
     pokok
    -0.41
     Exactos
    -0.41
     つ折り
    -0.40
    olidation
    -0.40
    ViewChild
    -0.39
    离
    -0.38
    shower
    -0.37
    Great
    -0.37
     tráng
    -0.37
    POSITIVE LOGITS
    mtrl
    0.44
    protoimpl
    0.43
    def
    0.43
    \{\\
    0.42
    attr
    0.41
    AddAttribute
    0.41
     DSL
    0.41
     ListTile
    0.40
    CppMethod
    0.40
     StatelessWidget
    0.39
    Activations Density 0.152%

    No Known Activations