© 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. 34-GEMMASCOPE-RES-16K
    6. 11338
    Prev
    Next
    INDEX
    Explanations

    quantitative measures of weight or volume in relation to resources

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_34/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.34.hook_resid_post
    Hook Layer
    34
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     soie
    -0.43
     artificiales
    -0.34
     chão
    -0.34
     Tracks
    -0.33
     encaje
    -0.32
    IndentedString
    -0.32
     Gedicht
    -0.32
     marbre
    -0.32
    zieren
    -0.31
     aparcamiento
    -0.31
    POSITIVE LOGITS
     ton
    1.58
     tonnes
    1.53
     tons
    1.50
     tonne
    1.47
     Tons
    1.34
     Ton
    1.34
    Ton
    1.30
    Tons
    1.28
     TON
    1.26
     toneladas
    1.25
    Activations Density 0.316%

    No Known Activations