© 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-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1861
    Prev
    Next
    INDEX
    Explanations

    code snippets and programming constructs related to error handling and signal processing

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

    low-level system programming code, specifically code related to system signals and error handling in Go.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    code syntax elements like operators, punctuation, and control structures in programming languages.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
     dolayı
    -0.38
    intios
    -0.38
     my
    -0.37
     thank
    -0.36
    неопр
    -0.36
    Rüyada
    -0.35
    cerpt
    -0.34
     dynamic
    -0.34
    sa
    -0.34
    évaluateur
    -0.34
    POSITIVE LOGITS
    UnsafeEnabled
    0.68
     ProtoMessage
    0.67
     protoimpl
    0.66
    protoimpl
    0.64
    tangentMode
    0.52
    codecSelfer
    0.51
     yyhl
    0.50
    adpleegd
    0.50
     intStringLen
    0.50
    AuxInt
    0.48
    Activations Density 0.897%

    No Known Activations