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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 4-GEMMASCOPE-RES-16K
    6. 3445
    Prev
    Next
    INDEX
    Explanations

    URL paths and content-related elements in code syntax

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WriteTagHelper
    -0.69
     ModelExpression
    -0.67
     colonia
    -0.64
     gereken
    -0.61
    StructEnd
    -0.58
    QMetaType
    -0.53
    expandindo
    -0.53
    GeneratedMessage
    -0.53
     üzere
    -0.52
    istocene
    -0.51
    POSITIVE LOGITS
    TestingModule
    0.62
     незавершена
    0.60
     disambiguazione
    0.56
    زع
    0.54
     vak
    0.53
    gonic
    0.53
    Czytaj
    0.52
    WithMany
    0.51
    elfth
    0.50
    tarea
    0.49
    Activations Density 0.027%

    No Known Activations