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. 20-GEMMASCOPE-RES-16K
    6. 407
    Prev
    Next
    INDEX
    Explanations

    terms related to programming instructions and CPU operations

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

    Code, instructions, or expressions

    np_acts-logits-general · gemini-2.0-flash

    call or statement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     الرخصة
    -0.66
     насељу
    -0.57
     تضيفلها
    -0.56
    endix
    -0.53
    libft
    -0.53
    SuppressMessage
    -0.51
    OrNil
    -0.50
    sidemargin
    -0.50
    RetentionPolicy
    -0.49
     snippetHide
    -0.49
    POSITIVE LOGITS
     call
    1.09
     calls
    1.08
    calls
    1.00
    call
    1.00
     Calls
    0.91
    Call
    0.90
    Calls
    0.86
     CALL
    0.86
     Call
    0.85
    CALL
    0.82
    Activations Density 0.639%

    No Known Activations