© 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. 2-GEMMASCOPE-RES-16K
    6. 1734
    Prev
    Next
    INDEX
    Explanations

    certain conjunctions and expressions indicating relationships between elements in a sentence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     accessible
    -0.60
    ouchable
    -0.55
    bable
    -0.54
     web
    -0.54
     utafitiHapana
    -0.54
    lecte
    -0.54
     estimés
    -0.54
    ujących
    -0.53
    OKAY
    -0.53
    ̣i
    -0.52
    POSITIVE LOGITS
     estekak
    0.64
     createState
    0.62
    SharedCtor
    0.59
    GHIJKLM
    0.59
    uxxxx
    0.59
    AndEndTag
    0.57
     onlyOwner
    0.57
    awaiter
    0.57
    DockStyle
    0.55
    migrationBuilder
    0.54
    Activations Density 0.491%

    No Known Activations