© 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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 3493
    Prev
    Next
    INDEX
    Explanations

    terms related to binding and constraints in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Tokens related to "bind" and "binding"

    np_acts-logits-general · gemini-2.0-flash

    bind, binding, bound, binds

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     betweenstory
    -0.72
    abestanden
    -0.62
    ValueStyle
    -0.59
    +#+
    -0.56
    </s>
    -0.53
    tosh
    -0.51
    ⃗
    -0.50
    odeon
    -0.50
    حياته
    -0.50
    endpush
    -0.49
    POSITIVE LOGITS
     Bind
    1.36
     binding
    1.36
     Binding
    1.32
     binds
    1.32
     BIND
    1.30
     bind
    1.29
     bindings
    1.16
    binding
    1.16
     Bound
    1.13
     bound
    1.09
    Activations Density 0.123%

    No Known Activations