© 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. 37-GEMMASCOPE-RES-16K
    6. 3908
    Prev
    Next
    INDEX
    Explanations

    references to personal aspirations and future plans

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.37.hook_resid_post
    Hook Layer
    37
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    已被
    -0.33
    JspWriter
    -0.31
     chapéu
    -0.29
     xícara
    -0.28
     geçir
    -0.28
    乓
    -0.28
    Besos
    -0.28
     boîte
    -0.27
    kuh
    -0.27
    indd
    -0.27
    POSITIVE LOGITS
     gonna
    1.99
     going
    1.95
     GONNA
    1.49
    gonna
    1.45
     Gonna
    1.45
    Gonna
    1.38
     GOING
    1.33
    Going
    1.30
    going
    1.25
     Going
    1.19
    Activations Density 0.267%

    No Known Activations