© 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. 35-GEMMASCOPE-RES-16K
    6. 7468
    Prev
    Next
    INDEX
    Explanations

    medical terms and procedures related to health and treatment

    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
    fluffy
    -0.46
     casada
    -0.45
     dzie
    -0.45
     zagran
    -0.44
     fidanz
    -0.44
     محفوظة
    -0.43
    าะ
    -0.43
    uchtung
    -0.43
     GlobalKey
    -0.42
     actualidad
    -0.42
    POSITIVE LOGITS
     Савезне
    0.68
     transférez
    0.59
    bibfield
    0.57
     transfieras
    0.55
     houſe
    0.52
    httphttps
    0.49
     Houſe
    0.48
    المكان
    0.48
    enderror
    0.47
     '\\;'
    0.47
    Activations Density 3.157%

    No Known Activations