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-2-2B
    4. Residual Stream - 16k
    5. 5-GEMMASCOPE-RES-16K
    6. 15565
    Prev
    Next
    INDEX
    Explanations

    the word "however" and its variations, indicating contrast or exception

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Lend
    -0.81
     Morton
    -0.71
     Dalton
    -0.68
    ziesz
    -0.63
     Alb
    -0.63
    Rite
    -0.63
     Rite
    -0.62
     Fight
    -0.62
     brunt
    -0.62
     burn
    -0.62
    POSITIVE LOGITS
     however
    1.67
     However
    1.62
    however
    1.61
    However
    1.52
     entanto
    1.39
     Porém
    1.28
    theless
    1.26
     Cependant
    1.18
     Tuttavia
    1.17
    Cependant
    1.10
    Activations Density 0.061%

    No Known Activations