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. 22-GEMMASCOPE-RES-16K
    6. 10885
    Prev
    Next
    INDEX
    Explanations

    programmatic structures and definitions in a code or scripting context

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

    Code, file paths, and software installation

    np_acts-logits-general · gemini-2.0-flash

    main_x64 components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    gmail
    -0.59
     CreateTagHelper
    -0.54
    WriteBarrier
    -0.52
     StatelessWidget
    -0.51
    standes
    -0.50
     $^{
    -0.50
    ниципа
    -0.49
     noDo
    -0.48
     mật
    -0.48
     tartalomajánló
    -0.47
    POSITIVE LOGITS
    _
    2.16
    \_
    1.18
     _
    1.09
    _{
    0.87
    _-_
    0.82
    }_
    0.80
    _$
    0.79
    _<
    0.73
    _%
    0.72
    '_
    0.71
    Activations Density 0.671%

    No Known Activations