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. 17-GEMMASCOPE-RES-16K
    6. 10671
    Prev
    Next
    INDEX
    Explanations

    closing brackets and indentation levels in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    /"
    -0.49
     للاسماء
    -0.47
    '/',
    -0.45
     getopt
    -0.45
    ::<
    -0.45
    cstdio
    -0.44
    вань
    -0.44
    HOY
    -0.43
     %>
    -0.42
    cobra
    -0.42
    POSITIVE LOGITS
    else
    1.66
     else
    1.62
     Else
    1.16
     ELSE
    1.10
     otherwise
    1.09
    Otherwise
    1.06
     Otherwise
    1.05
    Else
    1.01
    otherwise
    1.00
     others
    0.95
    Activations Density 0.047%

    No Known Activations