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. 35-GEMMASCOPE-RES-16K
    6. 14141
    Prev
    Next
    INDEX
    Explanations

    Java annotations and imports related to configuration and beans

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     informée
    -0.57
    UrlResolution
    -0.52
    principalColumn
    -0.52
    EndContext
    -0.52
    \{\\
    -0.50
    abestanden
    -0.47
    PerformLayout
    -0.47
    ########.
    -0.46
    RTLR
    -0.45
    errHandler
    -0.44
    POSITIVE LOGITS
     annotation
    1.27
     annotations
    1.07
    annotation
    1.05
     Annotation
    1.03
     annot
    1.02
    注解
    1.02
    annotations
    1.01
    annot
    0.93
     annotated
    0.91
    Annotation
    0.86
    Activations Density 0.230%

    No Known Activations