© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 6978
    Prev
    Next
    INDEX
    Explanations

    terms related to engineering processes and inspections

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Optimal
    -0.57
     dodgy
    -0.51
     hugely
    -0.51
     sobretudo
    -0.50
     Optimal
    -0.49
    Firstly
    -0.49
    optimal
    -0.48
    HasAnnotation
    -0.48
     optimised
    -0.48
     tweaked
    -0.47
    POSITIVE LOGITS
     shall
    0.71
    shall
    0.56
     Shall
    0.54
     wyma
    0.54
     erforderlich
    0.52
     SHALL
    0.52
     hereinafter
    0.51
     pož
    0.49
     anzuzeigen
    0.49
     objectionable
    0.48
    Activations Density 0.987%

    No Known Activations