© 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. 4840
    Prev
    Next
    INDEX
    Explanations

    technical instructions and references related to software or systems configuration

    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
     informée
    -0.61
     HasFactory
    -0.54
    checkFor
    -0.47
     esternos
    -0.44
     invokingState
    -0.43
    Diweddarwch
    -0.43
     Numerade
    -0.42
    expandindo
    -0.42
    providedIn
    -0.41
     pourrez
    -0.40
    POSITIVE LOGITS
     staffs
    0.80
     equipments
    0.75
     quitted
    0.71
     firstly
    0.71
     recommand
    0.68
    
    0.63
     sth
    0.63
    Ⅱ
    0.63
     stuffs
    0.62
    
    0.61
    Activations Density 5.074%

    No Known Activations