Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 1223
    Prev
    Next
    INDEX
    Explanations

    legal terminology and procedural phrases

    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
    Four
    -0.44
    FOUR
    -0.42
    four
    -0.35
     Four
    -0.35
    انتهای
    -0.30
     Fore
    -0.29
    Fore
    -0.29
     four
    -0.29
    fourth
    -0.28
    fore
    -0.28
    POSITIVE LOGITS
     Италијани
    0.61
     XLV
    0.59
    rungsseite
    0.57
    
    0.57
     forty
    0.57
    tvguidetime
    0.56
     Forty
    0.56
    adaptiveStyles
    0.55
    principalColumn
    0.55
    ngrx
    0.54
    Activations Density 1.578%

    No Known Activations