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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 14389
    Prev
    Next
    INDEX
    Explanations

    phrases related to subscription or signing up

    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
    )]:
    -0.57
     استنادى
    -0.56
    iyle
    -0.52
    ohistochemistry
    -0.49
    )":
    -0.49
    fum
    -0.49
    }';
    -0.48
     Directives
    -0.48
    hdashline
    -0.48
     Lade
    -0.47
    POSITIVE LOGITS
     registration
    1.64
     registrations
    1.54
     registering
    1.53
     Registration
    1.52
     registered
    1.47
     register
    1.46
    Registration
    1.36
     Register
    1.36
     Regist
    1.36
    Regist
    1.33
    Activations Density 0.140%

    No Known Activations