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. 18-GEMMASCOPE-RES-16K
    6. 2267
    Prev
    Next
    INDEX
    Explanations

    numerical values and their representations

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

    Numbers between 1 and 9

    np_acts-logits-general · gemini-2.0-flash

    numbers or numerical sequences

    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
    AnchorTagHelper
    -0.83
     CYTHON
    -0.81
    HasForeignKey
    -0.79
    +#+
    -0.76
     Waray
    -0.73
    ":[{
    -0.73
     surla
    -0.71
    BagLayout
    -0.71
    }));
    
    -0.71
    CodedInputStream
    -0.70
    POSITIVE LOGITS
     oxalate
    0.61
     useHistory
    0.60
     TNA
    0.59
    addVertex
    0.55
     défend
    0.54
     trå
    0.53
     défendre
    0.52
    ethene
    0.52
    callee
    0.51
    ガニック
    0.51
    Activations Density 0.521%

    No Known Activations